In order to be widely applicable, speech-driven 3D head avatars must articulate their lips in accordance with speech, while also conveying the appropriate emotions with dynamically changing facial expressions. The key problem is that deterministic models produce high-quality lip-sync but without rich expressions, whereas stochastic models generate diverse expressions but with lower lip-sync quality. To get the best of both, we seek a stochastic model with accurate lipsync. To that end, we develop a new approach based on the following observation: if a method generates realistic 3D lip motions, it should be possible to infer the spoken audio from the lip motion. The inferred speech should match the original input audio, and erroneous predictions create a novel supervision signal for training 3D talking head avatars with accurate lip-sync. To demonstrate this effect, we propose THUNDER (Talking Heads Under Neural Differentiable Elocution Reconstruction), a 3D talking head avatar framework that introduces a novel supervision mechanism via differentiable sound production. First, we train a novel mesh-tospeech model that regresses audio from facial animation. Then, we incorporate this model into a diffusion-based talking avatar framework. During training, the mesh-to-speech model takes the generated animation and produces a sound that is compared to the input speech, creating a differentiable analysis-by-audio-synthesis supervision loop. Our extensive qualitative and quantitative experiments demonstrate that THUNDER significantly improves the quality of the lip-sync of talking head avatars while still allowing for generation of diverse, high-quality, expressive facial animations.

Reconstructing realistic animal fur geometry from images is a challenging task due to the fine-scale details, selfocclusion, and view-dependent appearance of fur. In contrast to human hairstyle reconstruction, there are also no datasets that can be leveraged to learn a fur prior for different animals. In this work, we present a first multi-viewbased method for high-fidelity 3D fur modeling of animals using a strand-based representation, leveraging the general knowledge of a vision language model. Given multiview RGB images, we first reconstruct a coarse surface geometry using traditional multi-view stereo techniques. We then use a vision language model (VLM) system to retrieve information about the realistic length structure of the fur for each part of the body. We use this knowledge to construct the animal's furless geometry and grow strands atop it. The fur reconstruction is supervised with both geometric and photometric losses computed from multi-view images. To mitigate orientation ambiguities stemming from the Gabor filters that are applied to the input images, we additionally utilize the VLM to guide the strands' growth direction and their relation to the gravity vector that we incorporate as a loss. With this new schema of using a VLM to guide 3D reconstruction from multi-view inputs, we show generalization across a variety of animals with different fur types. For additional results and code, please refer to https://neuralfur.is.tue.mpg.de.

We present HaPTIC, an approach that infers coherent 4D hand trajectories from monocular videos. Current videobased hand pose reconstruction methods primarily focus on improving frame-wise 3D pose using adjacent frames rather than studying consistent 4D hand trajectories in space. Despite the additional temporal cues, they generally underperform compared to image-based methods due to the scarcity of annotated video data. To address these issues, we repurpose a state-of-the-art image-based transformer to take in multiple frames and directly predict a coherent trajectory. We introduce two types of lightweight attention layers: cross-view self-attention to fuse temporal information, and global cross-attention to bring in larger spatial context. Our method infers 4D hand trajectories similar to the ground truth while maintaining strong 2 D reprojection alignment. We apply the method to both egocentric and allocentric videos. It significantly outperforms existing methods in global trajectory accuracy while being comparable to the state-of-the-art in single-image pose estimation.

We address the challenge of accurate 3D human pose and shape estimation from monocular images. The key to accuracy and robustness lies in high-quality training data. Existing training datasets containing real images with pseudo ground truth (pGT) use SMPLify to fit SMPL to sparse 2D joint locations, assuming a simplified camera with default intrinsics. We make two contributions that improve pGT accuracy. First, to estimate camera intrinsics, we develop a field-of-view prediction model (HumanFoV) trained on a dataset of images containing people. We use the estimated intrinsics to enhance the 4D-Humans dataset by incorporating a full perspective camera model during SMPLify fitting. Second, 2D joints provide limited constraints on 3D body shape, resulting in average-looking bodies. To address this, we use the BEDLAM dataset to train a dense surface keypoint detector. We apply this detector to the 4D-Humans dataset and modify SMPLify to fit the detected keypoints, resulting in significantly more realistic body shapes. Finally, we upgrade the HMR2.0 architecture to include the estimated camera parameters. We iterate model training and SMPLify fitting initialized with the previously trained model. This leads to more accurate pGT and a new model, CameraHMR, with state-of-the-art accuracy. Code and pGT is available for research purposes.

We introduce Gaussian Garments, a novel approach for reconstructing realistic simulation-ready garment assets from multi-view videos. Our method represents garments with a combination of a 3D mesh and a Gaussian texture that encodes both the color and high-frequency surface details. This representation enables accurate registration of garment geometries to multi-view videos and helps disentangle albedo textures from lighting effects. Furthermore, we demonstrate how a pretrained graph neural network (GNN) can be fine-tuned to replicate the real behavior of each garment. The reconstructed Gaussian Garments can be automatically combined into multi-garment outfits and animated with the fine-tuned GNN.

Predicting the dynamics of interacting objects is essential for both humans and intelligent systems. However, existing approaches are limited to simplified, toy settings and lack generalizability to complex, real-world environments. Recent advances in generative models have enabled the prediction of state transitions based on interventions, but focus on generating a single future state which neglects the continuous dynamics resulting from the interaction. To address this gap, we propose InterDyn, a novel framework that generates videos of interactive dynamics given an initial frame and a control signal encoding the motion of a driving object or actor. Our key insight is that large video generation models can act as both neural renderers and implicit physics "simulators", having learned interactive dynamics from large-scale video data. To effectively harness this capability, we introduce an interactive control mechanism that conditions the video generation process on the motion of the driving entity. Qualitative results demonstrate that InterDyn generates plausible, temporally consistent videos of complex object interactions while generalizing to unseen objects. Quantitative evaluations show that InterDyn outperforms baselines that focus on static state transitions. This work highlights the potential of leveraging video generative models as implicit physics engines. Code and trained models will be released at: https://interdyn.is.tue.mpg.de/.

We introduce ChatGarment, a novel approach that leverages large vision-language models (VLMs) to automate the estimation, generation, and editing of 3D garments from images or text descriptions. Unlike previous methods that struggle in real-world scenarios or lack interactive editing capabilities, ChatGarment can estimate sewing patterns from in-the-wild images or sketches, generate them from text descriptions, and edit garments based on user instructions, all within an interactive dialogue. These sewing patterns can then be draped on a 3D body and animated. This is achieved by finetuning a VLM to directly generate a JSON file that includes both textual descriptions of garment types and styles, as well as continuous numerical attributes. This JSON file is then used to create sewing patterns through a programming parametric model. To support this, we refine the existing programming model, GarmentCode, by expanding its garment type coverage and simplifying its structure for efficient VLM fine-tuning. Additionally, we construct a large-scale dataset of image-to-sewing-pattern and text-to-sewing-pattern pairs through an automated data pipeline. Extensive evaluations demonstrate ChatGarment’s ability to accurately reconstruct, generate, and edit garments from multimodal inputs, highlighting its potential to simplify work-flows in fashion and gaming applications. Code and data are available at https://chatgarment.github.io/.

Recovering 3D Human-Object Interaction (HOI) from single images is challenging due to depth ambiguities, occlusions, and the huge variation in object shape and appearance. Thus, past work requires controlled settings such as known object shapes and contacts, and tackles only limited object classes. Instead, we need methods that generalize to natural images and novel object classes. We tackle this in two main ways: (1) We collect PICO-db, a new dataset of natural images uniquely paired with dense 3D contact correspondences on both body and object meshes. To this end, we use images from the recent DAMON dataset that are paired with annotated contacts, but only on a canonical 3D body. In contrast, we seek contact labels on both the body and the object. To infer these, given an image, we retrieve an appropriate 3D object mesh from a database by leveraging vision foundation models. Then, we project DAMON’s body contact patches onto the object via a novel method needing only 2 clicks per patch. This minimal human input establishes rich contact correspondences between bodies and objects. (2) We exploit our new dataset in a novel render-and-compare fitting method, called PICO-fit, to recover 3D body and object meshes in interaction. PICO-fit infers contact for the SMPL-X body, retrieves a likely 3D object mesh and contact from PICO-db for that object, and uses the contact to iteratively fit the 3D body and object meshes to image evidence via optimization. Uniquely, PICO-fit works well for many object classes that no existing method can tackle. This is crucial for scaling HOI understanding in the wild. Our data and code are available at https://pico.is.tue.mpg.de.

We introduce InteractVLM, a novel method to estimate 3D contact points on human bodies and objects from single in- the-wild images, enabling accurate human-object joint reconstruction in 3D. This is challenging due to occlusions, depth ambiguities, and widely varying object shapes. Existing methods rely on 3D contact annotations collected via expensive motion-capture systems or tedious manual labeling, limiting scalability and generalization. To overcome this, InteractVLM harnesses the broad visual knowledge of large Vision-Language Models (VLMs), fine-tuned with limited 3D contact data. However, directly applying these models is non-trivial, as they reason "only" in 2D, while human-object contact is inherently 3D. Thus we introduce a novel "Render-Localize-Lift" module that: (1) embeds 3D body and object surfaces in 2D space via multiview rendering, (2) trains a novel multi-view localization model (MV-Loc) to infer contacts in 2D, and (3) lifts these to 3D. Additionally, we propose a new task called Semantic Human Contact estimation, where human contact predictions are conditioned explicitly on object semantics, enabling richer interaction modeling. InteractVLM outperforms existing work on contact estimation and also facilitates 3D reconstruction from an in-the-wild image. To estimate 3D human and object pose, we infer initial body and object meshes, then infer contacts on both of these via InteractVLM, and last exploit these for fitting the meshes to image evidence. Results show that our approach performs promisingly in the wild. Code and models are available at https://interactvlm.is.tue.mpg.de.

The notion of 3D reconstruction as scene understanding is foundational in computer vision. Reconstructing 3D scenes from 2D visual observations necessitates strong priors to disambiguate structure. Much work has been focused on the anthropocentric, which, characterized by smooth surfaces, coherent normals, and regular edges, allows for the integration of strong geometric inductive biases. Here, we consider a more challenging problem where such assumptions do not hold: the reconstruction of natural scenes containing trees, bushes, boulders, and animals. While numerous works have attempted to tackle the problem of reconstructing animals in the wild, they have focused solely on the animal, neglecting environmental context. This limits their usefulness for analysis tasks, as animals exist inherently within the 3D world, and information is lost when environmental factors are disregarded. We propose a method to reconstruct natural scenes from single images. We base our approach on recent advances leveraging the strong world priors ingrained in Large Language Models and train an autoregressive model to decode a CLIP embedding into a structured, compositional scene representation, encompassing both animals and the wild. To enable this, we propose a synthetic dataset comprising one million images and thousands of assets. Through our introduction of a CLIP-projection head, we demonstrate that our approach generalizes to the task of reconstructing animals and their environments in real-world images, despite having been trained solely on synthetic data. We release our dataset and code to encourage future research at https://raw.is.tue.mpg.de/.

Numerous methods have been proposed to detect, estimate, and analyze properties of people in images, including 3D pose, shape, contact, human-object interaction, and emotion. While widely applicable in vision and other areas, such methods require expert knowledge to select, use, and interpret the results. To address this, we introduce ChatHuman, a language-driven system that integrates the capabilities of specialized methods into a unified framework. ChatHuman functions as an assistant proficient in utilizing, analyzing, and interacting with tools specific to 3D human tasks, adeptly discussing and resolving related challenges. Built on a Large Language Model (LLM) framework, ChatHuman is trained to autonomously select, apply, and interpret a diverse set of tools in response to user inputs. Our approach overcomes significant hurdles in adapting LLMs to 3D human tasks, including the need for domain-specific knowledge and the ability to interpret complex 3D outputs. The innovations of ChatHuman include leveraging academic publications to instruct the LLM on tool usage, employing a retrieval-augmented generation model to create in-context learning examples for managing new tools, and effectively discriminating between and integrating tool results by transforming specialized 3D outputs into comprehensible formats. Experiments demonstrate that ChatHuman surpasses existing models in both tool selection accuracy and overall performance across various 3D human tasks, and it supports interactive chatting with users. ChatHuman represents a significant step toward consolidating diverse analytical methods into a unified, robust system for 3D human tasks. Code and data are available at chathuman.github.io.

We address the task of generating 3D hair geometry from a single image, which is challenging due to the diversity of hairstyles and the lack of paired image-to-3D hair data. Previous methods are primarily trained on synthetic data and cope with the limited amount of such data by using low-dimensional intermediate representations, such as guide strands and scalp-level embeddings, that require post-processing to decode, upsample, and add realism. These approaches fail to reconstruct detailed hair, struggle with curly hair, or are limited to handling only a few hairstyles. To overcome these limitations, we propose DiffLocks, a novel framework that enables detailed reconstruction of a wide variety of hairstyles directly from a single image. First, we address the lack of 3D hair data by automating the creation of the largest synthetic hair dataset to date, containing 40K hairstyles. Second, we leverage the synthetic hair dataset to learn an image-conditioned diffusion-transfomer model that generates accurate 3D strands from a single frontal image. By using a pretrained image backbone, our method generalizes to in-the-wild images despite being trained only on synthetic data. Our diffusion model predicts a scalp texture map in which any point in the map contains the latent code for an individual hair strand. These codes are directly decoded to 3D strands without post-processing techniques. Representing individual strands, instead of guide strands, enables the transformer to model the detailed spatial structure of complex hairstyles. With this, DiffLocks can recover highly curled hair, like afro hairstyles, from a single image for the first time. Data and code is available at https://radualexandru.github.io/difflocks/.

Human pose and shape (HPS) estimation presents challenges in diverse scenarios such as crowded scenes, person-person interactions, and single-view reconstruction. Existing approaches lack mechanisms to incorporate auxiliary "side information" that could enhance reconstruction accuracy in such challenging scenarios. Furthermore, the most accurate methods rely on cropped person detections and cannot exploit scene context while methods that process the whole image often fail to detect people and are less accurate than methods that use crops. While recent language-based methods explore HPS reasoning through large language or vision-language models, their metric accuracy is well below the state of the art. In contrast, we present PromptHMR, a transformer-based promptable method that reformulates HPS estimation through spatial and semantic prompts. Our method processes full images to maintain scene context and accepts multiple input modalities: spatial prompts like bounding boxes and masks, and semantic prompts like language descriptions or interaction labels. PromptHMR demonstrates robust performance across challenging scenarios: estimating people from bounding boxes as small as faces in crowded scenes, improving body shape estimation through language descriptions, modeling person-person interactions, and producing temporally coherent motions in videos. Experiments on benchmarks show that PromptHMR achieves state-of-the-art performance while offering flexible prompt-based control over the HPS estimation process.

We present MoGA, a novel method to reconstruct high-fidelity 3D Gaussian avatars from a single-view image. The main challenge lies in inferring unseen appearance and geometric details while ensuring 3D consistency and realism. Most previous methods rely on 2D diffusion models to synthesize unseen views; however, these generated views are sparse and inconsistent, resulting in unrealistic 3D artifacts and blurred appearance. To address these limitations, we leverage a generative avatar model, that can generate diverse 3D avatars by sampling deformed Gaussians from a learned prior distribution. Due to limited 3D training data, such a 3D model alone cannot capture all image details of unseen identities. Consequently, we integrate it as a prior, ensuring 3D consistency by projecting input images into its latent space and enforcing additional 3D appearance and geometric constraints. Our novel approach formulates Gaussian avatar creation as model inversion by fitting the generative avatar to synthetic views from 2D diffusion models. The generative avatar provides an initialization for model fitting, enforces 3D regularization, and helps in refining pose. Experiments show that our method surpasses state-of-the-art techniques and generalizes well to real-world scenarios. Our Gaussian avatars are also inherently animatable. For code, see https://zj-dong.github.io/MoGA/.

Dynamic 3D reconstruction and point tracking in videos are typically treated as separate tasks, despite their deep connection. We propose St4RTrack, a feed-forward framework that simultaneously reconstructs and tracks dynamic video content in a world coordinate frame from RGB inputs. This is achieved by predicting two appropriately defined pointmaps for a pair of frames captured at different moments. Specifically, we predict both pointmaps at the same moment, in the same world, capturing both static and dynamic scene geometry while maintaining 3D correspondences. Chaining these predictions through the video sequence with respect to a reference frame naturally computes long-range correspondences, effectively combining 3D reconstruction with 3D tracking. Unlike prior methods that rely heavily on 4D ground truth supervision, we employ a novel adaptation scheme based on a reprojection loss. We establish a new extensive benchmark for world-frame reconstruction and tracking, demonstrating the effectiveness and efficiency of our unified, data-driven framework. Our code, model, and benchmark will be released.

Capturing accurate 3D human pose in the wild would provide valuable data for training pose estimation and motion generation methods. While video-based estimation approaches have become increasingly accurate, they often fail in common scenarios involving self-contact, such as a hand touching the face. In contrast, wearable bioimpedance sensing can cheaply and unobtrusively measure ground-truth skin-to-skin contact. Consequently, we propose a novel framework that combines visual pose estimators with bioimpedance sensing to capture the 3D pose of people by taking self-contact into account. Our method, BioTUCH, initializes the pose using an off-the-shelf estimator and introduces contact-aware pose optimization during measured self-contact: reprojection error and deviations from the input estimate are minimized while enforcing vertex proximity constraints. We validate our approach using a new dataset of synchronized RGB video, bioimpedance measurements, and 3D motion capture. Testing with three input pose estimators, we demonstrate an average of 11.7% improvement in reconstruction accuracy. We also present a miniature wearable bioimpedance sensor that enables efficient large-scale collection of contact-aware training data for improving pose estimation and generation using BioTUCH. Code and data are available at biotuch.is.tue.mpg.de

Fitting a body to a 3D clothed human point cloud is a common yet challenging task. Traditional optimization-based approaches use multi-stage pipelines that are sensitive to pose initialization, while recent learning-based methods often struggle with generalization across diverse poses and garment types. We propose Equivariant Tightness Fitting for Clothed Humans, or ETCH, a novel pipeline that estimates cloth-to-body surface mapping through locally approximate SE(3) equivariance, encoding tightness as displacement vectors from the cloth surface to the underlying body. Following this mapping, pose-invariant body features regress sparse body markers, simplifying clothed human fitting into an inner-body marker fitting task. Extensive experiments on CAPE and 4D-Dress show that ETCH significantly outperforms state-of-the-art methods -- both tightness-agnostic and tightness-aware -- in body fitting accuracy on loose clothing (16.7% ~ 69.5%) and shape accuracy (average 49.9%). Our equivariant tightness design can even reduce directional errors by (67.2% ~ 89.8%) in one-shot (or out-of-distribution) settings (~ 1% data). Qualitative results demonstrate strong generalization of ETCH, regardless of challenging poses, unseen shapes, loose clothing, and non-rigid dynamics. We will release the code and models soon for research purposes at https://boqian-li.github.io/ETCH/.

We formulate the motor system of an interactive avatar as a generative motion model that can drive the body to move through 3D space in a perpetual, realistic, controllable, and responsive manner. Although human motion generation has been extensively studied, many existing methods lack the responsiveness and realism of real human movements. Inspired by recent advances in foundation models, we propose PRIMAL, which is learned with a two-stage paradigm. In the pretraining stage, the model learns body movements from a large number of sub-second motion segments, providing a generative foundation from which more complex motions are built. This training is fully unsupervised without annotations. Given a single-frame initial state during inference, the pretrained model not only generates unbounded, realistic, and controllable motion, but also enables the avatar to be responsive to induced impulses in real time. In the adaptation phase, we employ a novel ControlNet-like adaptor to fine-tune the base model efficiently, adapting it to new tasks such as few-shot personalized action generation and spatial target reaching. Evaluations show that our proposed method outperforms state-of-the-art baselines. We leverage the model to create a real-time character animation system in Unreal Engine that feels highly responsive and natural. Code, models, and more results are available at: https://yz-cnsdqz.github.io/eigenmotion/PRIMAL

Against the backdrop of enthusiasm for large language models (LLMs), there is a growing need to scientifically assess their capabilities and shortcomings. This is nontrivial in part because it is difficult to find tasks which the models have not encountered during training. Utilizing symbolic graphics programs, we propose a domain well-suited to test multiple spatial-semantic reasoning skills of LLMs. Popular in computer graphics, these programs procedurally generate visual data. While LLMs exhibit impressive skills in general program synthesis and analysis, symbolic graphics programs offer a new layer of evaluation: they allow us to test an LLM's ability to answer semantic questions about the images or 3D geometries without a vision encoder. To semantically understand the symbolic programs, LLMs would need to possess the ability to "imagine" and reason how the corresponding graphics content would look with only the symbolic description of the local curvatures and strokes. We use this task to evaluate LLMs by creating a large benchmark for the semantic visual understanding of symbolic graphics programs, built procedurally with minimal human effort. Particular emphasis is placed on transformations of images that leave the image level semantics invariant while introducing significant changes to the underlying program. We evaluate commercial and open-source LLMs on our benchmark to assess their ability to reason about visual output of programs, finding that LLMs considered stronger at reasoning generally perform better. Lastly, we introduce a novel method to improve this ability -- Symbolic Instruction Tuning (SIT), in which the LLM is finetuned with pre-collected instruction data on symbolic graphics programs. Interestingly, we find that SIT not only improves LLM's understanding on symbolic programs, but it also improves general reasoning ability on various other benchmarks.

Synthesizing high-quality 3D head textures is crucial for gaming, virtual reality, and digital humans. Achieving seamless 360° textures typically requires expensive multi-view datasets with precise tracking. However, traditional methods struggle without back-view data or precise geometry, especially for human heads, where even minor inconsistencies disrupt realism. We introduce HairFree, an unsupervised texturing framework guided by textual descriptions and 2D diffusion priors, producing high-consistency 360° bald head textures—including non-human skin with fine details—without any texture, back-view, bald, non-human, or synthetic training data. We fine-tune a diffusion prior on a dataset of mostly frontal faces, conditioned on predicted 3D head geometry and face parsing. During inference, HairFree uses precise skin masks and 3D FLAME geometry as input conditioning, ensuring high 3D consistency and alignment. We synthesize the full 360° texture by first generating a frontal RGB image aligned to the 3D FLAME pose and mapping it to UV space. As the virtual camera moves, we inpaint and merge missing regions. A built-in semantic prior enables precise region separation—particularly for isolating and removing hair—allowing seamless integration with various assets like customizable 3D hair, eyeglasses, jewelry, etc. We evaluate HairFree quantitatively and qualitatively, demonstrating its superiority over state-of-the-art 3D head avatar generation methods. https://hairfree.is.tue.mpg.de/

Inferring 3D human motion from video remains a challenging problem with many applications. While traditional methods estimate the human in image coordinates, many applications require human motion to be estimated in world coordinates. This is particularly challenging when there is both human and camera motion. Progress on this topic has been limited by the lack of rich video data with ground truth human and camera movement. We address this with BEDLAM2.0, a new dataset that goes beyond the popular BEDLAM dataset in important ways. In addition to introducing more diverse and realistic cameras and camera motions, BEDLAM2.0 increases diversity and realism of body shape, motions, clothing, hair, and 3D environments. Additionally, it adds shoes, which were missing in BEDLAM. BEDLAM has become a key resource for training 3D human pose and motion regressors today and we show that BEDLAM2.0 is significantly better, particularly for training methods that estimate humans in world coordinates. We compare state-of-the art methods trained on BEDLAM and BEDLAM2.0, and find that BEDLAM2.0 significantly improves accuracy over BEDLAM. For research purposes, we provide the rendered videos, ground truth body parameters, and camera motions. We also provide the 3D assets to which we have rights and links to those from third parties.

Manipulating the illumination of a 3D scene within a single image represents a fundamental challenge in computer vision and graphics. This problem has traditionally been addressed using inverse rendering techniques, which involve explicit 3D asset reconstruction and costly ray-tracing simulations. Meanwhile, recent advancements in visual foundation models suggest that a new paradigm could soon be possible – one that replaces explicit physical models with networks that are trained on large amounts of image and video data. In this paper, we exploit the implicit scene understanding of a video diffusion model, particularly Stable Video Diffusion, to relight a single image. We introduce GenLit, a framework that distills the ability of a graphics engine to perform light manipulation into a video-generation model, enabling users to directly insert and manipulate a point light in the 3D world within a given image and generate results directly as a video sequence. We find that a model fine-tuned on only a small synthetic dataset generalizes to real-world scenes, enabling single-image relighting with plausible and convincing shadows and inter-reflections. Our results highlight the ability of video foundation models to capture rich information about lighting, material, and shape, and our findings indicate that such models, with minimal training, can be used to perform relighting without explicit asset reconstruction or ray-tracing.

Pose estimation has promised to impact healthcare by enabling more practical methods to quantify nuances of human movement and biomechanics. However, despite the inherent connection between pose estimation and biomechanics, these disciplines have largely remained disparate. For example, most current pose estimation benchmarks use metrics such as Mean Per Joint Position Error, Percentage of Correct Keypoints, or mean Average Precision to assess performance, without quantifying kinematic and physiological correctness - key aspects for biomechanics. To alleviate this challenge, we develop OpenCapBench to offer an easy-to-use unified benchmark to assess common tasks in human pose estimation, evaluated under physiological constraints. OpenCapBench computes consistent kinematic metrics through joints angles provided by an open-source musculoskeletal modeling software (OpenSim). Through OpenCapBench, we demonstrate that current pose estimation models use keypoints that are too sparse for accurate biomechanics analysis. To mitigate this challenge, we introduce SynthPose, a new approach that enables finetuning of pre-trained 2D human pose models to predict an arbitrarily denser set of keypoints for accurate kinematic analysis through the use of synthetic data. Incorporating such fine-tuning on synthetic data of prior models leads to twofold reduced joint angle errors. Moreover, OpenCapBench allows users to benchmark their own developed models on our clinically relevant cohort. Overall, OpenCapBench bridges the computer vision and biomechanics communities, aiming to drive simultaneous advances in both areas.

The regression of 3D Human Pose and Shape (HPS) from an image is becoming increasingly accurate. This makes the results useful for downstream tasks like human action recognition or 3D graphics. Yet, no regressor is perfect, and accuracy can be affected by ambiguous image evidence or by poses and appearance that are unseen during training. Most current HPS regressors, however, do not report the confidence of their outputs, meaning that downstream tasks cannot differentiate accurate estimates from inaccurate ones. To address this, we develop POCO, a novel framework for training HPS regressors to estimate not only a 3D human body, but also their confidence, in a single feed-forward pass. Specifically, POCO estimates both the 3D body pose and a per-sample variance. The key idea is to introduce a Dual Conditioning Strategy (DCS) for regressing uncertainty that is highly correlated to pose reconstruction quality. The POCO framework can be applied to any HPS regressor and here we evaluate it by modifying HMR, PARE, and CLIFF. In all cases, training the network to reason about uncertainty helps it learn to more accurately estimate 3D pose. While this was not our goal, the improvement is modest but consistent. Our main motivation is to provide uncertainty estimates for downstream tasks; we demonstrate this in two ways: (1) We use the confidence estimates to bootstrap HPS training. Given unlabeled image data, we take the confident estimates of a POCO-trained regressor as pseudo ground truth. Retraining with this automatically-curated data improves accuracy. (2) We exploit uncertainty in video pose estimation by automatically identifying uncertain frames (e.g. due to occlusion) and “inpainting” these from confident frames. Code and models are available for research at https://poco.is.tue.mpg.de.

We present a method to estimate human motion in a global scene from moving cameras. This is a highly challenging task due to the coupling of human and camera motions in the video. To address this problem, we propose a joint optimization framework that disentangles human and camera motions using both foreground human motion priors and background scene features. Unlike existing methods that use SLAM as initialization, we propose to tightly integrate SLAM and human motion priors in an optimization that is inspired by bundle adjustment. Specifically, we optimize human and camera motions to match both the observed human pose and scene features. This design combines the strengths of SLAM and motion priors, which leads to significant improvements in human and camera motion estimation. We additionally introduce a motion prior that is suitable for batch optimization, making our approach significantly more efficient than existing approaches. Finally, we propose a novel synthetic dataset that enables evaluating camera motion in addition to human motion from dynamic videos. Experiments on the synthetic and real-world RICH datasets demonstrate that our approach substantially outperforms prior art in recovering both human and camera motions.

We introduce TADA, a simple-yet-effective approach that takes textual descriptions and produces expressive 3D avatars with high-quality geometry and lifelike textures, that can be animated and rendered with traditional graphics pipelines. Existing text-based character generation methods are limited in terms of geometry and texture quality, and cannot be realistically animated due to the misalignment between the geometry and the texture, particularly in the face region. To address these limitations, TADA leverages the synergy of a 2D diffusion model and a parametric body model. Specifically, we derive a high-resolution upsampled version of SMPL-X with a displacement layer and a texture map, and use hierarchical rendering with score distillation sampling (SDS) to create high-quality, detailed, holistic 3D avatars from text. To ensure alignment between the geometry and texture, we render normals and RGB images of the generated character and exploit their latent embeddings during the SDS optimization process. We further drive the character’s face with multiple expressions during optimization, ensuring that its semantics remain consistent with the original SMPL-X model. Both qualitative and quantitative evaluations show that TADA significantly surpasses existing approaches. TADA enables large-scale creation of digital characters ready for animation and rendering, while also enabling text-guided editing. The code is public for research purposes at tada.is.tue.mpg.de

Hands are dexterous and highly versatile manipulators that are central to how humans interact with objects and their environment. Consequently, modeling realistic hand-object interactions, including the subtle motion of individual fingers, is critical for applications in computer graphics, computer vision, and mixed reality. Prior work on capturing and modeling humans interacting with objects in 3D focuses on the body and object motion, often ignoring hand pose. In contrast, we introduce GRIP, a learning-based method that takes, as input, the 3D motion of the body and the object, and synthesizes realistic motion for both hands before, during, and after object interaction. As a preliminary step before synthesizing the hand motion, we first use a network, ANet, to denoise the arm motion. Then, we leverage the spatio-temporal relationship between the body and the object to extract novel temporal interaction cues, and use them in a two-stage inference pipeline to generate the hand motion. In the first stage, we introduce a new approach to encourage motion temporal consistency in the latent space (LTC) and generate consistent interaction motions. In the second stage, GRIP generates refined hand poses to avoid hand-object penetrations. Given sequences of noisy body and object motion, GRIP “upgrades” them to include hand-object interaction. Quantitative experiments and perceptual studies demonstrate that GRIP outperforms baseline methods and generalizes to unseen objects and motions from different motion-capture datasets. Our models and code are available for research purposes at https://grip.is.tue.mpg.de.

Our goal is to create a realistic 3D facial avatar with hair and accessories using only a text description. While this challenge has attracted significant recent interest, existing methods either lack realism, produce unrealistic shapes, or do not support editing, such as modifications to the hairstyle. We argue that existing methods are limited because they employ a monolithic modeling approach, using a single representation for the head, face, hair, and accessories. Our observation is that the hair and face, for example, have very different structural qualities that benefit from different representations. Building on this insight, we generate avatars with a compositional model, in which the head, face, and upper body are represented with traditional 3D meshes, and the hair, clothing, and accessories with neural radiance fields (NeRF). The model-based mesh representation provides a strong geometric prior for the face region, improving realism while enabling editing of the person’s appearance. By using NeRFs to represent the remaining components, our method is able to model and synthesize parts with complex geometry and appearance, such as curly hair and fluffy scarves. Our novel system synthesizes these high-quality compositional avatars from text descriptions. Specifically, we generate a face image using text, fit a parametric shape model to it, and inpaint texture using diffusion models. Conditioned on the generated face, we sequentially generate style components such as hair or clothing using Score Distillation Sampling (SDS) with guidance from CLIPSeg segmentations. However, this alone is not sufficient to produce avatars with a high degree of realism. Consequently, we introduce a hierarchical approach to refine the non-face regions using a BLIP-based loss combined with SDS. The experimental results demonstrate that our method, Text-guided generation and Editing of Compositional Avatars (TECA), produces avatars that are more realistic than those of recent methods while being editable because of their compositional nature. For example, our TECA enables the seamless transfer of compositional features like hairstyles, scarves, and other accessories between avatars. This capability supports applications such as virtual try-on. The code and generated avatars will be publicly available for research purposes at yfeng95.github.io/teca.

Existing methods for synthesizing 3D human gestures from speech have shown promising results, but they do not explicitly model the impact of emotions on the generated gestures. Instead, these methods directly output animations from speech without control over the expressed emotion. To address this limitation, we present AMUSE, an emotional speech-driven body animation model based on latent diffusion. Our observation is that content (i.e., gestures related to speech rhythm and word utterances), emotion, and personal style are separable. To account for this, AMUSE maps the driving audio to three disentangled latent vectors: one for content, one for emotion, and one for personal style. A latent diffusion model, trained to generate gesture motion sequences, is then conditioned on these latent vectors. Once trained, AMUSE synthesizes 3D human gestures directly from speech with control over the expressed emotions and style by combining the content from the driving speech with the emotion and style of another speech sequence. Randomly sampling the noise of the diffusion model further generates variations of the gesture with the same emotional expressivity. Qualitative, quantitative, and perceptual evaluations demonstrate that AMUSE outputs realistic gesture sequences. Compared to the state of the art, the generated gestures are better synchronized with the speech content, and better represent the emotion expressed by the input speech. Our code is available at amuse.is.tue.mpg.de.

Synthesizing natural human motions that enable a 3D human avatar to walk and reach for arbitrary goals in 3D space remains an unsolved problem with many applications. Existing methods (data-driven or using reinforcement learning) are limited in terms of generalization and motion naturalness. A primary obstacle is the scarcity of training data that combines locomotion with goal reaching. To ad-dress this, we introduce WANDR, a data-driven model that takes an avatar's initial pose and a goal's 3D position and generates natural human motions that place the end effec-tor (wrist) on the goal location. To solve this, we intro-duce novel intention features that drive rich goal-oriented movement. Intention guides the agent to the goal, and in-teractively adapts the generation to novel situations without needing to define sub-goals or the entire motion path. Cru-cially, intention allows training on datasets that have goal-oriented motions as well as those that do not. WANDR is a conditional Variational Auto-Encoder (c- VAE), which we train using the AMASS and CIRCLE datasets. We evaluate our method extensively and demonstrate its ability to gener-ate natural and long-term motions that reach 3D goals and generalize to unseen goal locations. Our models and code are available for research purposes at wandr.is.tue.mpg.de.

We address the problem of regressing 3D human pose and shape from a single image, with a focus on 3D accuracy. The current best methods leverage large datasets of 3D pseudo-ground-truth (p-GT) and 2D keypoints, leading to robust performance. With such methods, however, we observe a paradoxical decline in 3D pose accuracy with increasing 2D accuracy. This is caused by biases in the p-GT and the use of an approximate camera projection model. We quantify the error induced by current camera models and show that fitting 2D keypoints and p-GT accurately causes incorrect 3D poses. Our analysis defines the invalid distances within which minimizing 2D and p-GT losses is detrimental. We use this to formulate a new loss, “Threshold-Adaptive Loss Scaling” (TALS), that penalizes gross 2D and p-GT errors but not smaller ones. With such a loss, there are many 3D poses that could equally explain the 2D evidence. To reduce this ambiguity we need a prior over valid human poses but such priors can introduce unwanted bias. To address this, we exploit a tokenized representation of human pose and reformulate the problem as token prediction. This restricts the estimated poses to the space of valid poses, effectively improving robustness to occlusion. Exten-sive experiments on the EMDB and 3DPW datasets show that our reformulated loss and tokenization allows us to train on in-the-wild data while improving 3D accuracy over the state-of-the-art. Our models and code are available for research at https://tokenhmr.is.tue.mpg.de.

Since humans interact with diverse objects every day, the holistic 3D capture of these interactions is important to understand and model human behaviour. However, most ex-isting methods for hand-object reconstruction from RGB ei-ther assume pre-scanned object templates or heavily rely on limited 3D hand-object data, restricting their ability to scale and generalize to more unconstrained interaction settings. To address this, we introduce HOLD the first category-agnostic method that reconstructs an articulated hand and an object jointly from a monocular interaction video. We develop a compositional articulated implicit model that can reconstruct disentangled 3D hands and ob-jects from 2D images. We also further incorporate hand-object constraints to improve hand-object poses and con-sequently the reconstruction quality. Our method does not rely on any 3D hand-object annotations while significantly outperforming fully-supervised baselines in both in-the-lab and challenging in-the-wild settings. Moreover, we qualita-tively show its robustness in reconstructing from in-the-wild videos. See here for code, data, models, and updates.

We introduce ChatPose, a framework employing Large Language Models (LLMs) to understand and reason about 3D human poses from images or textual descriptions. Our work is motivated by the human ability to intuitively under-stand postures from a single image or a brief description, a process that intertwines image interpretation, world knowl-edge, and an understanding of body language. Traditional human pose estimation and generation methods often op-erate in isolation, lacking semantic understanding and rea-soning abilities. ChatPose addresses these limitations by embedding SMPL poses as distinct signal tokens within a multimodal LLM, enabling the direct generation of 3D body poses from both textual and visual inputs. Leveraging the powerful capabilities of multimodal LLMs, ChatPose uni-fies classical 3D human pose and generation tasks while offering user interactions. Additionally, ChatPose empow-ers LLMs to apply their extensive world knowledge in rea-soning about human poses, leading to two advanced tasks: speculative pose generation and reasoning about pose esti-mation. These tasks involve reasoning about humans to generate 3D poses from subtle text queries, possibly ac-companied by images. We establish benchmarks for these tasks, moving beyond traditional 3D pose generation and estimation methods. Our results show that ChatPose out-performs existing multimodal LLMs and task-specific meth-ods on these newly proposed tasks. Furthermore, Chat-Pose's ability to understand and generate 3D human poses based on complex reasoning opens new directions in human pose analysis. Code and data are available for research at https://yfeng95.github.io/ChatPose.

The creation of personalized anatomical digital twins is important in the fields of medicine, computer graphics, sports science, and biomechanics. To observe a subject's anatomy, expensive medical devices (MRI or CT) are required and the creation of the digital model is often time-consuming and involves manual effort. Instead, we leverage the fact that the shape of the body surface is correlated with the internal anatomy; e.g. from surface observations alone, one can predict body composition and skeletal structure. In this work, we go further and learn to infer the 3D location of three important anatomic tissues: subcutaneous adipose tissue (fat), lean tissue (muscles and organs), and long bones. To learn to infer these tissues, we tackle several key challenges. We first create a dataset of human tissues by segmenting full-body MRI scans and registering the SMPL body mesh to the body surface. With this dataset, we train HIT (Human Implicit Tissues), an implicit function that, given a point inside a body, predicts its tissue class. HIT leverages the SMPL body model shape and pose parameters to canonicalize the medical data. Unlike SMPL, which is trained from upright 3D scans, MRI scans are acquired with subjects lying on a table, resulting in significant soft-tissue deformation. Consequently, HIT uses a learned volumetric deformation field that undoes these deformations. Since HIT is parameterized by SMPL, we can repose bodies or change the shape of subjects and the internal structures deform appropriately. We perform extensive experiments to validate HIT's ability to predict a plausible internal structure for novel subjects. The dataset and HIT model are available at https://hit.is.tue.mpg.de to foster future research in this direction.

We propose EMAGE, a framework to generate full-body human gestures from audio and masked gestures, encompassing facial, local body, hands, and global movements. To achieve this, we first introduce BEAT2 (BEAT-SMPLX-FLAME), a new mesh-level holistic co-speech dataset. BEAT2 combines a MoShed SMPL-X body with FLAME head parameters and further refines the modeling of head, neck, and finger movements, offering a community-standardized, high-quality 3D motion captured dataset. EMAGE leverages masked body gesture priors during training to boost inference performance. It involves a Masked Audio Gesture Transformer, facilitating joint training on audio-to-gesture generation and masked gesture reconstruction to effectively encode audio and body gesture hints. Encoded body hints from masked gestures are then separately employed to generate facial and body movements. Moreover, EMAGE adaptively merges speech features from the audio's rhythm and content and utilizes four compositional VQ-VAEs to enhance the results' fidelity and diversity. Experiments demonstrate that EMAGE generates holistic gestures with state-of-the-art performance and is flexible in accepting predefined spatial-temporal gesture inputs, generating complete, audio-synchronized results. Our code and dataset are available.1

Social interaction is a fundamental aspect of human behavior and communication. The way individuals position themselves in relation to others, also known as proxemics, conveys social cues and affects the dynamics of social interaction. Reconstructing such interaction from images presents challenges because of mutual occlusion and the limited availability of large training datasets. To address this, we present a novel approach that learns a prior over the 3D proxemics two people in close social interaction and demonstrate its use for single-view 3D reconstruction. We start by creating 3D training data of interacting people using image datasets with contact annotations. We then model the proxemics using a novel denoising diffusion model called BUDDI that learns the joint distribution over the poses of two people in close social interaction. Sampling from our generative proxemics model produces realistic 3D human interactions, which we validate through a perceptual study. We use BUDDI in reconstructing two people in close proximity from an image without any contact annotation via an optimization approach that uses the diffusion model as a prior. Our approach recovers accurate 3D social interactions from noisy initial estimates, outperforming state-of-the-art methods. Our code, data, and model are available at: muelea. github.io/buddi.

We present SCULPT, a novel 3D generative model for clothed and textured 3D meshes of humans. Specifically, we devise a deep neural network that learns to represent the geometry and appearance distribution of clothed human bodies. Training such a model is challenging, as datasets of textured 3D meshes for humans are limited in size and accessibility. Our key observation is that there exist medium-sized 3D scan datasets like CAPE, as well as large-scale 2D image datasets of clothed humans and multiple appearances can be mapped to a single geometry. To effectively learn from the two data modalities, we propose an unpaired learning procedure for pose-dependent clothed and textured human meshes. Specifically, we learn a pose-dependent geometry space from 3D scan data. We represent this as per vertex displacements w.r.t. the SMPL model. Next, we train a geometry conditioned texture generator in an unsupervised way using the 2D image data. We use intermediate activations of the learned geometry model to condition our texture generator. To alleviate entanglement between pose and clothing type, and pose and clothing appearance, we condition both the texture and geometry generators with attribute labels such as clothing types for the geometry, and clothing colors for the texture generator. We automatically generated these conditioning labels for the 2D images based on the visual question answering model BLIP and CLIP. We validate our method on the SCULPT dataset, and compare to state-of-the-art 3D generative models for clothed human bodies. Our code and data can be found at https://sculpt.is.tue.mpg.de.

The estimation of 3D human motion from video has progressed rapidly but current methods still have several key limitations. First, most methods estimate the human in camera coordinates. Second, prior work on estimating humans in global coordinates often assumes a flat ground plane and produces foot sliding. Third, the most accurate methods rely on computationally expensive optimization pipelines, limiting their use to offline applications. Finally, existing video-based methods are surprisingly less accurate than singleframe methods. We address these limitations with WHAM (World-grounded Humans with Accurate Motion), which accurately and efficiently reconstructs 3D human motion in a global coordinate system from video. WHAM learns to lift 2D keypoint sequences to 3D using motion capture data and fuses this with video features, integrating motion context and visual information. WHAM exploits camera angular velocity estimated from a SLAM method together with human motion to estimate the body's global trajectory. We combine this with a contact-aware trajectory refinement method that lets WHAM capture human motion in diverse conditions, such as climbing stairs. WHAM outperforms all existing 3D human motion recovery methods across multiple in-the-wild benchmarks. Code is available for research purposes at http://wham.is.tue.mpg.del.

We present HAAR, a new strand-based generative model for 3D human hairstyles. Specifically, based on textual inputs, HAAR produces 3D hairstyles that could be used as production-level assets in modern computer graphics engines. Current AI-based generative models take advantage of powerful 2D priors to reconstruct 3D content in the form of point clouds, meshes, or volumetric functions. However, by using the 2D priors, they are intrinsically limited to only recovering the visual parts. Highly occluded hair structures can not be reconstructed with those methods, and they only model the “outer shell”, which is not ready to be used in physics-based rendering or simulation pipelines. In contrast, we propose a first text-guided generative method that uses 3D hair strands as an underlying representation. Leveraging 2D visual question-answering (VQA) systems, we automatically annotate synthetic hair models that are generated from a small set of artist-created hairstyles. This allows us to train a latent diffusion model that operates in a common hairstyle UV space. In qualitative and quantitative studies, we demonstrate the capabilities of the proposed model and compare it to existing hairstyle generation approaches. For results, please refer to our project page.

Data-driven three-dimensional parametric shape mod-els of the human body have gained enormous popularity both for the analysis of visual data and for the generation of synthetic humans. Following a similar approach for animals does not scale to the multitude of existing ani-mal species, not to mention the difficulty of accessing sub-jects to scan in 3D. However, we argue that for domestic species of great importance, like the horse, it is a highly valuable investment to put effort into gathering a large dataset of real 3D scans, and learn a realistic 3D articu-lated shape model. We introduce VAREN, a novel 3D ar-ticulated parametric shape model learned from 3D scans of many real horses. VAREN bridges synthesis and analysis tasks, as the generated model instances have unprecedented realism, while being able to represent horses of different sizes and shapes. Differently from previous body models, VAREN has two resolutions, an anatomical skeleton, and interpretable, learned pose-dependent deformations, which are related to the body muscles. We show with experiments that this formulation has superior performance with respect to previous strategies for modeling pose-dependent deformations in the human body case, while also being more compact and allowing an analysis of the relationship be-tween articulation and muscle deformation during articu-lated motion. The VAREN model and data are available at https://varen.is.tue.mpg.de.

The creation of photorealistic virtual worlds requires the accurate modeling of 3D surface geometry for a wide range of objects. For this, meshes are appealing since they enable 1) fast physics-based rendering with realistic material and lighting, 2) physical simulation, and 3) are memory-efficient for modern graphics pipelines. Recent work on reconstructing and statistically modeling 3D shape, however, has critiqued meshes as being topologically inflexible. To capture a wide range of object shapes, any 3D representation must be able to model solid, watertight, shapes as well as thin, open, surfaces. Recent work has focused on the former, and methods for reconstructing open surfaces do not support fast reconstruction with material and lighting or unconditional generative modelling. Inspired by the observation that open surfaces can be seen as islands floating on watertight surfaces, we parametrize open surfaces by defining a manifold signed distance field on watertight templates. With this parametrization, we further develop a grid-based and differentiable representation that parametrizes both watertight and non-watertight meshes of arbitrary topology. Our new representation, called Ghost-on-the-Shell (G-Shell), enables two important applications: differentiable rasterization-based reconstruction from multiview images and generative modelling of non-watertight meshes. We empirically demonstrate that G-Shell achieves state-of-the-art performance on non-watertight mesh reconstruction and generation tasks, while also performing effectively for watertight meshes.

Large foundation models are becoming ubiquitous, but training them from scratch is prohibitively expensive. Thus, efficiently adapting these powerful models to downstream tasks is increasingly important. In this paper, we study a principled finetuning paradigm -- Orthogonal Finetuning (OFT) -- for downstream task adaptation. Despite demonstrating good generalizability, OFT still uses a fairly large number of trainable parameters due to the high dimensionality of orthogonal matrices. To address this, we start by examining OFT from an information transmission perspective, and then identify a few key desiderata that enable better parameter-efficiency. Inspired by how the Cooley-Tukey fast Fourier transform algorithm enables efficient information transmission, we propose an efficient orthogonal parameterization using butterfly structures. We apply this parameterization to OFT, creating a novel parameter-efficient finetuning method, called Orthogonal Butterfly (BOFT). By subsuming OFT as a special case, BOFT introduces a generalized orthogonal finetuning framework. Finally, we conduct an extensive empirical study of adapting large vision transformers, large language models, and text-to-image diffusion models to various downstream tasks in computer vision and natural language. The results validate the effectiveness of BOFT as a generic finetuning method.

Generative Adversarial Networks (GANs) can produce high-quality samples, but do not provide an estimate of the probability density around the samples. However, it has been noted that maximizing the log-likelihood within an energy-based setting can lead to an adversarial framework where the discriminator provides unnormalized density (often called energy). We further develop this perspective, incorporate importance sampling, and show that 1) Wasserstein GAN performs a biased estimate of the partition function, and we propose instead to use an unbiased estimator; and 2) when optimizing for likelihood, one must maximize generator entropy. This is hypothesized to provide a better mode coverage. Different from previous works, we explicitly compute the density of the generated samples. This is the key enabler to designing an unbiased estimator of the partition function and computation of the generator entropy term. The generator density is obtained via a new type of flow network, called one-way flow network, that is less constrained in terms of architecture, as it does not require a tractable inverse function. Our experimental results show that our method converges faster, produces comparable sample quality to GANs with similar architecture, successfully avoids over-fitting to commonly used datasets and produces smooth low-dimensional latent representations of the training data.

Understanding how humans interact with the world necessitates accurate 3D hand pose estimation, a task complicated by the hand’s high degree of articulation, frequent occlusions, self-occlusions, and rapid motions. While most existing methods rely on single-image inputs, videos have useful cues to address aforementioned issues. However, existing video-based 3D hand datasets are insufficient for training feedforward models to generalize to in-the-wild scenarios. On the other hand, we have access to large human motion capture datasets which also include hand motions, e.g. AMASS. Therefore, we develop a generative motion prior specific for hands, trained on the AMASS dataset which features diverse and high-quality hand motions. This motion prior is then employed for video-based 3D hand motion estimation following a latent optimization approach. Our integration of a robust motion prior significantly enhances performance, especially in occluded scenarios. It produces stable, temporally consistent results that surpass conventional single-frame methods. We demonstrate our method’s efficacy via qualitative and quantitative evaluations on the HO3D and DexYCB datasets, with special emphasis on an occlusion-focused subset of HO3D. Code is available at https://hmp.is.tue.mpg.de

Abstract Humans constantly interact with objects to accomplish tasks. To understand such interactions, computers need to reconstruct these in 3D from images of whole bodies manipulating objects, e.g., for grasping, moving and using the latter. This involves key challenges, such as occlusion between the body and objects, motion blur, depth ambiguities, and the low image resolution of hands and graspable object parts. To make the problem tractable, the community has followed a divide-and-conquer approach, focusing either only on interacting hands, ignoring the body, or on interacting bodies, ignoring the hands. However, these are only parts of the problem. On the contrary, recent work focuses on the whole problem. The GRAB dataset addresses whole-body interaction with dexterous hands but captures motion via markers and lacks video, while the BEHAVE dataset captures video of body-object interaction but lacks hand detail. We address the limitations of prior work with InterCap, a novel method that reconstructs interacting whole-bodies and objects from multi-view RGB-D data, using the parametric whole-body SMPL-X model and known object meshes. To tackle the above challenges, InterCap uses two key observations: (i) Contact between the body and object can be used to improve the pose estimation of both. (ii) Consumer-level Azure Kinect cameras let us set up a simple and flexible multi-view RGB-D system for reducing occlusions, with spatially calibrated and temporally synchronized cameras. With our InterCap method we capture the InterCap dataset, which contains 10 subjects (5 males and 5 females) interacting with 10 daily objects of various sizes and affordances, including contact with the hands or feet. To this end, we introduce a new data-driven hand motion prior, as well as explore simple ways for automatic contact detection based on 2D and 3D cues. In total, InterCap has 223 RGB-D videos, resulting in 67,357 multi-view frames, each containing 6 RGB-D images, paired with pseudo ground-truth 3D body and object meshes. Our InterCap method and dataset fill an important gap in the literature and support many research directions. Data and code are available at https://intercap.is.tue.mpg.de .

Generating personalized 3D avatars is crucial for AR/VR. However, recent text-to-3D methods that generate avatars for celebrities or fictional characters, struggle with everyday people. Methods for faithful reconstruction typically require full-body images in controlled settings. What if users could just upload their personal "OOTD" (Outfit Of The Day) photo collection and get a faithful avatar in return? The challenge is that such casual photo collections contain diverse poses, challenging viewpoints, cropped views, and occlusion (albeit with a consistent outfit, accessories and hairstyle). We address this novel " Album2Human " task by developing PuzzleAvatar , a novel model that generates a faithful 3D avatar (in a canonical pose) from a personal OOTD album, bypassing the challenging estimation of body and camera pose. To this end, we fine-tune a foundational vision-language model (VLM) on such photos, encoding the appearance, identity, garments, hairstyles, and accessories of a person into separate learned tokens, instilling these cues into the VLM. In effect, we exploit the learned tokens as "puzzle pieces" from which we assemble a faithful, personalized 3D avatar. Importantly, we can customize avatars by simply inter-changing tokens. As a benchmark for this new task, we create a new dataset, called PuzzleIOI , with 41 subjects in a total of nearly 1k OOTD configurations, in challenging partial photos with paired ground-truth 3D bodies. Evaluation shows that PuzzleAvatar not only has high reconstruction accuracy, outperforming TeCH and MVDreamBooth, but also a unique scalability to album photos, and demonstrating strong robustness. Our code and data are publicly available for research purpose at puzzleavatar.is.tue.mpg.de

We show, for the first time, that neural networks trained only on synthetic data achieve state-of-the-art accuracy on the problem of 3D human pose and shape (HPS) estimation from real images. Previous synthetic datasets have been small, unrealistic, or lacked realistic clothing. Achieving sufficient realism is non-trivial and we show how to do this for full bodies in motion. Specifically, our BEDLAM dataset contains monocular RGB videos with ground-truth 3D bodies in SMPL-X format. It includes a diversity of body shapes, motions, skin tones, hair, and clothing. The clothing is realistically simulated on the moving bodies using commercial clothing physics simulation. We render varying numbers of people in realistic scenes with varied lighting and camera motions. We then train various HPS regressors using BEDLAM and achieve state-of-the-art accuracy on real-image benchmarks despite training with synthetic data. We use BEDLAM to gain insights into what model design choices are important for accuracy. With good synthetic training data, we find that a basic method like HMR approaches the accuracy of the current SOTA method (CLIFF). BEDLAM is useful for a variety of tasks and all images, ground truth bodies, 3D clothing, support code, and more are available for research purposes. Additionally, we provide detailed information about our synthetic data generation pipeline, enabling others to generate their own datasets. See the project page: https://bedlam.is.tue.mpg.de/.

Existing methods for capturing datasets of 3D heads in dense semantic correspondence are slow and commonly address the problem in two separate steps; multi-view stereo (MVS) reconstruction followed by non-rigid registration. To simplify this process, we introduce TEMPEH (Towards Estimation of 3D Meshes from Performances of Expressive Heads) to directly infer 3D heads in dense correspondence from calibrated multi-view images. Registering datasets of 3D scans typically requires manual parameter tuning to find the right balance between accurately fitting the scans' surfaces and being robust to scanning noise and outliers. Instead, we propose to jointly register a 3D head dataset while training TEMPEH. Specifically, during training, we minimize a geometric loss commonly used for surface registration, effectively leveraging TEMPEH as a regularizer. Our multi-view head inference builds on a volumetric feature representation that samples and fuses features from each view using camera calibration information. To account for partial occlusions and a large capture volume that enables head movements, we use view-and surface-aware feature fusion, and a spatial transformer-based head localization module, respectively. We use raw MVS scans as supervision during training, but, once trained, TEMPEH directly predicts 3D heads in dense correspondence without requiring scans. Predicting one head takes about 0.3 seconds with a median reconstruction error of 0.26 mm, 64% lower than the current state-of-the-art. This enables the efficient capture of large datasets containing multiple people and diverse facial motions. Code, model, and data are publicly available at https://tempeh.is.tue.mpg.de.

Humans constantly contact objects to move and perform tasks. Thus, detecting human-object contact is important for building human-centered artificial intelligence. However, there exists no robust method to detect contact between the body and the scene from an image, and there exists no dataset to learn such a detector. We fill this gap with HOT (“Human-Object conTact”), a new dataset of human-object contacts in images. To build HOT, we use two data sources: (1) We use the PROX dataset of 3D human meshes moving in 3D scenes, and automatically annotate 2D image areas for contact via 3D mesh proximity and projection. (2) We use the V-COCO, HAKE and Watch-n-Patch datasets, and ask trained annotators to draw polygons around the 2D image areas where contact takes place. We also annotate the involved body part of the human body. We use our HOT dataset to train a new contact detector, which takes a single color image as input, and outputs 2D contact heatmaps as well as the body-part labels that are in contact. This is a new and challenging task, that extends current foot-ground or hand-object contact detectors to the full generality of the whole body. The detector uses a part-attention branch to guide contact estimation through the context of the surrounding body parts and scene. We evaluate our detector extensively, and quantitative results show that our model outperforms baselines, and that all components contribute to better performance. Results on images from an online repository show reasonable detections and generalizability. Our HOT data and model are available for research at https://hot.is.tue.mpg.de.

Humans intuitively understand that inanimate objects do not move by themselves, but that state changes are typically caused by human manipulation (e.g., the opening of a book). This is not yet the case for machines. In part this is because there exist no datasets with ground-truth 3D annotations for the study of physically consistent and synchronised motion of hands and articulated objects. To this end, we introduce ARCTIC - a dataset of two hands that dexterously manipulate objects, containing 2 .1M video frames paired with accurate 3D hand and object meshes and detailed, dynamic contact information. It contains bi-manual articulation of objects such as scissors or laptops, where hand poses and object states evolve jointly in time. We propose two novel articulated hand-object interaction tasks: (1) Consistent motion reconstruction: Given a monocular video, the goal is to reconstruct two hands and articulated objects in 3D, so that their motions are spatio-temporally consistent. (2) Interaction field estimation: Dense relative hand-object distances must be estimated from images. We introduce two baselines ArcticNet and InterField, respectively and evaluate them qualitatively and quantitatively on ARCTIC. Our code and data are available at https://arctic.is.tue.mpg.de.

Sign language (SL) is the primary method of communication for the 70 million Deaf people around the world. Video dictionaries of isolated signs are a core SL learning tool. Replacing these with 3D avatars can aid learning and enable AR/VR applications, improving access to technology and online media. However, little work has attempted to estimate expressive 3D avatars from SL video; occlusion, noise, and motion blur make this task difficult. We address this by introducing novel linguistic priors that are universally applicable to SL and provide constraints on 3D hand pose that help resolve ambiguities within isolated signs. Our method, SGNify, captures fine-grained hand pose, facial expression, and body movement fully automatically from in-the-wild monocular SL videos. We evaluate SGNify quantitatively by using a commercial motion-capture system to compute 3D avatars synchronized with monocular video. SGNify outperforms state-of-the-art 3D body-pose-and shape-estimation methods on SL videos. A perceptual study shows that SGNify's 3D reconstructions are significantly more comprehensible and natural than those of previous methods and are on par with the source videos. Code and data are available at sgnify.is.tue.mpg.de.

We propose a method that leverages graph neural networks, multi-level message passing, and unsupervised training to enable efficient prediction of realistic clothing dynamics. Whereas existing methods based on linear blend skinning must be trained for specific garments, our method, called HOOD, is agnostic to body shape and applies to tight-fitting garments as well as loose, free-flowing clothing. Furthermore, HOOD handles changes in topology (e.g., garments with buttons or zippers) and material properties at inference time. As one key contribution, we propose a hierarchical message-passing scheme that efficiently propagates stiff stretching modes while preserving local detail. We empirically show that HOOD outperforms strong baselines quantitatively and that its results are perceived as more realistic than state-of-the-art methods.

We address the problem of inferring the 3D shape and pose of dogs from images. Given the lack of 3D training data, this problem is challenging, and the best methods lag behind those designed to estimate human shape and pose. To make progress, we attack the problem from multiple sides at once. First, we need a good 3D shape prior, like those available for humans. To that end, we learn a dog-specific 3D parametric model, called D-SMAL. Second, existing methods focus on dogs in standing poses because when they sit or lie down, their legs are self occluded and their bodies deform. Without access to a good pose prior or 3D data, we need an alternative approach. To that end, we exploit contact with the ground as a form of side information. We consider an existing large dataset of dog images and label any 3D contact of the dog with the ground. We exploit body-ground contact in estimating dog pose and find that it significantly improves results. Third, we develop a novel neural network architecture to infer and exploit this contact information. Fourth, to make progress, we have to be able to measure it. Current evaluation metrics are based on 2D features like keypoints and silhouettes, which do not directly correlate with 3D errors. To address this, we create a synthetic dataset containing rendered images of scanned 3D dogs. With these advances, our method recovers significantly better dog shape and pose than the state of the art, and we evaluate this improvement in 3D. Our code, model and test dataset are publicly available for research purposes at https://bite.is.tue.mpg.de/.

Although the estimation of 3D human pose and shape (HPS) is rapidly progressing, current methods still cannot reliably estimate moving humans in global coordinates, which is critical for many applications. This is particularly challenging when the camera is also moving, entangling human and camera motion. To address these issues, we adopt a novel 5D representation (space, time, and identity) that enables end-to-end reasoning about people in scenes. Our method, called TRACE, introduces several novel architectural components. Most importantly, it uses two new “maps” to reason about the 3D trajectory of people over time in camera, and world, coordinates. An additional memory unit enables persistent tracking of people even during long occlusions. TRACE is the first one-stage method to jointly recover and track 3D humans in global coordinates from dynamic cameras. By training it end-to-end, and using full image information, TRACE achieves state-of-the-art performance on tracking and HPS benchmarks. The code11https://www.yusun.work/TRACE/TRACE.html and dataset22https://github.com/Arthur151/DynaCam are released for research purposes.

Estimating 3D humans from images often produces implausible bodies that lean, float, or penetrate the floor. Such methods ignore the fact that bodies are typically supported by the scene. A physics engine can be used to enforce physical plausibility, but these are not differentiable, rely on unrealistic proxy bodies, and are difficult to integrate into existing optimization and learning frameworks. In contrast, we exploit novel intuitive-physics (IP) terms that can be inferred from a 3D SMPL body interacting with the scene. Inspired by biomechanics, we infer the pressure heatmap on the body, the Center of Pressure (CoP) from the heatmap, and the SMPL body's Center of Mass (CoM). With these, we develop IPMAN, to estimate a 3D body from a color image in a “stable” configuration by encouraging plausible floor contact and overlapping CoP and CoM. Our IP terms are intuitive, easy to implement, fast to compute, differentiable, and can be integrated into existing optimization and regression methods. We evaluate IPMAN on standard datasets and MoYo, a new dataset with synchronized multi-view images, ground-truth 3D bodies with complex poses, body-floor contact, CoM and pressure. IPMAN produces more plausible results than the state of the art, improving accuracy for static poses, while not hurting dynamic ones. Code and data are available for research at https://ipman.is.tue.mpg.de.

The combination of deep learning, artist-curated scans, and Implicit Functions (IF), is enabling the creation of detailed, clothed, 3D humans from images. However, existing methods are far from perfect. IF-based methods recover free-form geometry, but produce disembodied limbs or degenerate shapes for novel poses or clothes. To increase robustness for these cases, existing work uses an explicit parametric body model to constrain surface reconstruction, but this limits the recovery of free-form surfaces such as loose clothing that deviates from the body. What we want is a method that combines the best properties of implicit representation and explicit body regularization. To this end, we make two key observations: (1) current networks are better at inferring detailed 2D maps than full-3D surfaces, and (2) a parametric model can be seen as a “canvas” for stitching together detailed surface patches. Based on these, our method, ECON, has three main steps: (1) It infers detailed 2D normal maps for the front and back side of a clothed person. (2) From these, it recovers 2.5D front and back surfaces, called d-BiNI, that are equally detailed, yet incomplete, and registers these w.r.t. each other with the help of a SMPL-X body mesh recovered from the image. (3) It “inpaints” the missing geometry between d-BiNI surfaces. If the face and hands are noisy, they can optionally be replaced with the ones of SMPL-X. As a result, ECON infers high-fidelity 3D humans even in loose clothes and challenging poses. This goes beyond previous methods, according to the quantitative evaluation on the CAPE and Renderpeople datasets. Perceptual studies also show that ECON's perceived realism is better by a large margin. Code and models are available for research purposes at econ.is.tue.mpg.de

Generating realistic 3D worlds occupied by moving humans has many applications in games, architecture, and synthetic data creation. But generating such scenes is expensive and labor intensive. Recent work generates human poses and motions given a 3D scene. Here, we take the opposite approach and generate 3D indoor scenes given 3D human motion. Such motions can come from archival motion capture or from IMU sensors worn on the body, effectively turning human movement into a “scanner” of the 3D world. Intuitively, human movement indicates the free-space in a room and human contact indicates surfaces or objects that support activities such as sitting, lying or touching. We propose MIME (Mining Interaction and Movement to infer 3D Environments), which is a generative model of indoor scenes that produces furniture layouts that are consistent with the human movement. MIME uses an auto-regressive transformer architecture that takes the already generated objects in the scene as well as the human motion as input, and outputs the next plausible object. To train MIME, we build a dataset by populating the 3D FRONT scene dataset with 3D humans. Our experiments show that MIME produces more diverse and plausible 3D scenes than a recent generative scene method that does not know about human movement. Code and data are available for research at https://mime.is.tue.mpg.de.

This work addresses the problem of generating 3D holistic body motions from human speech. Given a speech recording, we synthesize sequences of 3D body poses, hand gestures, and facial expressions that are realistic and diverse. To achieve this, we first build a high-quality dataset of 3D holistic body meshes with synchronous speech. We then define a novel speech-to-motion generation framework in which the face, body, and hands are modeled separately. The separated modeling stems from the fact that face articulation strongly correlates with human speech, while body poses and hand gestures are less correlated. Specifically, we employ an autoencoder for face motions, and a compositional vector-quantized variational autoencoder (VQ- VAE) for the body and hand motions. The compositional VQ-VAE is key to generating diverse results. Additionally, we propose a cross-conditional autoregressive model that generates body poses and hand gestures, leading to coherent and realistic motions. Extensive experiments and user studies demonstrate that our proposed approach achieves state-of-the-art performance both qualitatively and quantitatively. Our dataset and code are released for research purposes at https://talkshow.is.tue.mpg.de/.

The ability to create realistic animatable and relightable head avatars from casual video sequences would open up wide ranging applications in communication and entertainment. Current methods either build on explicit 3D morphable meshes (3DMM) or exploit neural implicit representations. The former are limited by fixed topology, while the latter are non-trivial to deform and inefficient to render. Furthermore, existing approaches entangle lighting and albedo, limiting the ability to re-render the avatar in new environments. In contrast, we propose PointAvatar, a deformable point-based representation that disentangles the source color into intrinsic albedo and normal-dependent shading. We demonstrate that PointAvatar bridges the gap between existing mesh- and implicit representations, combining high-quality geometry and appearance with topological flexibility, ease of deformation and rendering efficiency. We show that our method is able to generate animatable 3D avatars using monocular videos from multiple sources including hand-held smartphones, laptop webcams and internet videos, achieving state-of-the-art quality in challenging cases where previous methods fail, e.g., thin hair strands, while being significantly more efficient in training than competing methods.

Our goal is to synthesize 3D human motions given textual inputs describing simultaneous actions, for example ‘waving hand’ while ‘walking’ at the same time. We refer to generating such simultaneous movements as performing spatial compositions. In contrast to temporal compositions that seek to transition from one action to another, spatial compositing requires understanding which body parts are involved in which action, to be able to move them simultaneously. Motivated by the observation that the correspondence between actions and body parts is encoded in powerful language models, we extract this knowledge by prompting GPT-3 with text such as "what are the body parts involved in the action?", while also providing the parts list and few-shot examples. Given this action-part mapping, we combine body parts from two motions together and establish the first automated method to spatially compose two actions. However, training data with compositional actions is always limited by the combinatorics. Hence, we further create synthetic data with this approach, and use it to train a new state-of-the-art text-to-motion generation model, called SINC ("Simultaneous actioN Compositions for 3D human motions"). In our experiments, we find that training with such GPT-guided synthetic data improves spatial composition generation over baselines. Our code is publicly available at sinc.is.tue.mpg.de.

While progress in 2D generative models of human appearance has been rapid, many applications require 3D avatars that can be animated and rendered. Unfortunately, most existing methods for learning generative models of 3D humans with diverse shape and appearance require 3D training data, which is limited and expensive to acquire. The key to progress is hence to learn generative models of 3D avatars from abundant unstructured 2D image collections. However, learning realistic and complete 3D appearance and geometry in this under-constrained setting remains challenging, especially in the presence of loose clothing such as dresses. In this paper, we propose a new adversarial generative model of realistic 3D people learned from 2D images. Our method captures shape and deformation of the body and loose clothing by adopting a holistic 3D generator and integrating an efficient, flexible, articulation module. To improve realism, we train our model using multiple discriminators while also integrating geometric cues in the form of predicted 2D normal maps. We experimentally find that our method outperforms previous 3D- and articulation-aware methods in terms of geometry and appearance. We validate the effectiveness of our model and the importance of each component via systematic ablation studies.

We address the problem of fitting a parametric human body model (SMPL) to point cloud data. Optimization-based methods require careful initialization and are prone to becoming trapped in local optima. Learning-based methods address this but do not generalize well when the input pose is far from those seen during training. For rigid point clouds, remarkable generalization has been achieved by leveraging SE(3)-equivariant networks, but these methods do not work on articulated objects. In this work we extend this idea to human bodies and propose ArtEq, a novel part-based SE(3)-equivariant neural architecture for SMPL model estimation from point clouds. Specifically, we learn a part detection network by leveraging local SO(3) invariance, and regress shape and pose using articulated SE(3) shape-invariant and pose-equivariant networks, all trained end-to-end. Our novel pose regression module leverages the permutation-equivariant property of self-attention layers to preserve rotational equivariance. Experimental results show that ArtEq generalizes to poses not seen during training, outperforming state-of-the-art methods by ∼ 44% in terms of body reconstruction accuracy, without requiring an optimization refinement step. Furthermore, ArtEq is three orders of magnitude faster during inference than prior work and has 97.3% fewer parameters. The code and model are available for research purposes at https://arteq.is.tue.mpg.de.

In this paper, we present TMR, a simple yet effective approach for text to 3D human motion retrieval. While previous work has only treated retrieval as a proxy evaluation metric, we tackle it as a standalone task. Our method extends the state-of-the-art text-to-motion synthesis model TEMOS, and incorporates a contrastive loss to better structure the cross-modal latent space. We show that maintaining the motion generation loss, along with the contrastive training, is crucial to obtain good performance. We introduce a benchmark for evaluation and provide an in-depth analysis by reporting results on several protocols. Our extensive experiments on the KIT-ML and HumanML3D datasets show that TMR outperforms the prior work by a significant margin, for example reducing the median rank from 54 to 19. Finally, we showcase the potential of our approach on moment retrieval. Our code and models are publicly available at https://mathis.petrovich.fr/tmr.

Understanding how humans use physical contact to interact with the world is key to enabling human-centric artificial intelligence. While inferring 3D contact is crucial for modeling realistic and physically-plausible human-object interactions, existing methods either focus on 2D, consider body joints rather than the surface, use coarse 3D body regions, or do not generalize to in-the-wild images. In contrast, we focus on inferring dense, 3D contact between the full body surface and objects in arbitrary images. To achieve this, we first collect DAMON, a new dataset containing dense vertex-level contact annotations paired with RGB images containing complex human-object and human-scene contact. Second, we train DECO, a novel 3D contact detector that uses both body-part-driven and scene-context-driven attention to estimate vertex-level contact on the SMPL body. DECO builds on the insight that human observers recognize contact by reasoning about the contacting body parts, their proximity to scene objects, and the surrounding scene context. We perform extensive evaluations of our detector on DAMON as well as on the RICH and BEHAVE datasets. We significantly outperform existing SOTA methods across all benchmarks. We also show qualitatively that DECO generalizes well to diverse and challenging real-world human interactions in natural images. The code, data, and models are available at https://deco.is.tue.mpg.de.

In this paper, we focus on a general yet important learning problem, pairwise similarity learning (PSL). PSL subsumes a wide range of important applications, such as open-set face recognition, speaker verification, image retrieval and person re-identification. The goal of PSL is to learn a pairwise similarity function assigning a higher similarity score to positive pairs (i.e., a pair of samples with the same label) than to negative pairs (i.e., a pair of samples with different label). We start by identifying a key desideratum for PSL, and then discuss how existing methods can achieve this desideratum. We then propose a surprisingly simple proxy-free method, called SimPLE, which requires neither feature/proxy normalization nor angular margin and yet is able to generalize well in open-set recognition. We apply the proposed method to three challenging PSL tasks: open-set face recognition, image retrieval and speaker verification. Comprehensive experimental results on large-scale benchmarks show that our method performs significantly better than current state-of-the-art methods. Our project page is available at simple.is.tue.mpg.de.

To be widely adopted, 3D facial avatars must be animated easily, realistically, and directly from speech signals. While the best recent methods generate 3D animations that are synchronized with the input audio, they largely ignore the impact of emotions on facial expressions. Realistic facial animation requires lip-sync together with the natural expression of emotion. To that end, we propose EMOTE (Expressive Model Optimized for Talking with Emotion), which generates 3D talking-head avatars that maintain lip-sync from speech while enabling explicit control over the expression of emotion. To achieve this, we supervise EMOTE with decoupled losses for speech (i.e., lip-sync) and emotion. These losses are based on two key observations: (1) deformations of the face due to speech are spatially localized around the mouth and have high temporal frequency, whereas (2) facial expressions may deform the whole face and occur over longer intervals. Thus we train EMOTE with a per-frame lip-reading loss to preserve the speech-dependent content, while supervising emotion at the sequence level. Furthermore, we employ a content-emotion exchange mechanism in order to supervise different emotions on the same audio, while maintaining the lip motion synchronized with the speech. To employ deep perceptual losses without getting undesirable artifacts, we devise a motion prior in the form of a temporal VAE. Due to the absence of high-quality aligned emotional 3D face datasets with speech, EMOTE is trained with 3D pseudo-ground-truth extracted from an emotional video dataset (i.e., MEAD). Extensive qualitative and perceptual evaluations demonstrate that EMOTE produces speech-driven facial animations with better lip-sync than state-of-the-art methods trained on the same data, while offering additional, high-quality emotional control.

Abstract The goal of this work is to reconstruct 3D dogs from monocular images. We take a model-based approach, where we estimate the shape and pose parameters of a 3D articulated shape model for dogs. We consider dogs as they constitute a challenging problem, given they are highly articulated and come in a variety of shapes and appearances. Recent work has considered a similar task using the multi-animal SMAL model, with additional limb scale parameters, obtaining reconstructions that are limited in terms of realism. Like previous work, we observe that the original SMAL model is not expressive enough to represent dogs of many different breeds. Moreover, we make the hypothesis that the supervision signal used to train the network, that is 2D keypoints and silhouettes, is not sufficient to learn a regressor that can distinguish between the large variety of dog breeds. We therefore go beyond previous work in two important ways. First, we modify the SMAL shape space to be more appropriate for representing dog shape. Second, we formulate novel losses that exploit information about dog breeds. In particular, we exploit the fact that dogs of the same breed have similar body shapes. We formulate a novel breed similarity loss, consisting of two parts: One term is a triplet loss, that encourages the shape of dogs from the same breed to be more similar than dogs of different breeds. The second one is a breed classification loss. With our approach we obtain 3D dogs that, compared to previous work, are quantitatively better in terms of 2D reconstruction, and significantly better according to subjective and quantitative 3D evaluations. Our work shows that a-priori side information about similarity of shape and appearance, as provided by breed labels, can help to compensate for the lack of 3D training data. This concept may be applicable to other animal species or groups of species. We call our method BARC (Breed-Augmented Regression using Classification). Our code is publicly available for research purposes at https://barc.is.tue.mpg.de/ .

Neural fields have revolutionized the area of 3D reconstruction and novel view synthesis of rigid scenes. A key challenge in making such methods applicable to articulated objects, such as the human body, is to model the deformation of 3D locations between the rest pose (a canonical space) and the deformed space. We propose a new articulation module for neural fields, Fast-SNARF, which finds accurate correspondences between canonical space and posed space via iterative root finding. Fast-SNARF is a drop-in replacement in functionality to our previous work, SNARF, while significantly improving its computational efficiency. We contribute several algorithmic and implementation improvements over SNARF, yielding a speed-up of 150×. These improvements include voxel-based correspondence search, pre-computing the linear blend skinning function, and an efficient software implementation with CUDA kernels. Fast-SNARF enables efficient and simultaneous optimization of shape and skinning weights given deformed observations without correspondences (e.g. 3D meshes). Because learning of deformation maps is a crucial component in many 3D human avatar methods and since Fast-SNARF provides a computationally efficient solution, we believe that this work represents a significant step towards the practical creation of 3D virtual humans.

Our goal is to efficiently learn personalized animatable 3D head avatars from videos that are geometrically accurate, realistic, relightable, and compatible with current rendering systems. While 3D meshes enable efficient processing and are highly portable, they lack realism in terms of shape and appearance. Neural representations, on the other hand, are realistic but lack compatibility and are slow to train and render. Our key insight is that it is possible to efficiently learn high-fidelity 3D mesh representations via differentiable rendering by exploiting highly-optimized methods from traditional computer graphics and approximating some of the components with neural networks. To that end, we introduce FLARE, a technique that enables the creation of animatable and relightable mesh avatars from a single monocular video. First, we learn a canonical geometry using a mesh representation, enabling efficient differentiable rasterization and straightforward animation via learned blendshapes and linear blend skinning weights. Second, we follow physically-based rendering and factor observed colors into intrinsic albedo, roughness, and a neural representation of the illumination, allowing the learned avatars to be relit in novel scenes. Since our input videos are captured on a single device with a narrow field of view, modeling the surrounding environment light is non-trivial. Based on the split-sum approximation for modeling specular reflections, we address this by approximating the prefiltered environment map with a multi-layer perceptron (MLP) modulated by the surface roughness, eliminating the need to explicitly model the light. We demonstrate that our mesh-based avatar formulation, combined with learned deformation, material, and lighting MLPs, produces avatars with high-quality geometry and appearance, while also being efficient to train and render compared to existing approaches.

Great progress has been made in estimating 3D human pose and shape from images and video by training neural networks to directly regress the parameters of parametric human models like SMPL. However, existing body models have simplified kinematic structures that do not correspond to the true joint locations and articulations in the human skeletal system, limiting their potential use in biomechanics. On the other hand, methods for estimating biomechanically accurate skeletal motion typically rely on complex motion capture systems and expensive optimization methods. What is needed is a parametric 3D human model with a biomechanically accurate skeletal structure that can be easily posed. To that end, we develop SKEL, which re-rigs the SMPL body model with a biomechanics skeleton. To enable this, we need training data of skeletons inside SMPL meshes in diverse poses. We build such a dataset by optimizing biomechanically accurate skeletons inside SMPL meshes from AMASS sequences. We then learn a regressor from SMPL mesh vertices to the optimized joint locations and bone rotations. Finally, we re-parametrize the SMPL mesh with the new kinematic parameters. The resulting SKEL model is animatable like SMPL but with fewer, and biomechanically-realistic, degrees of freedom. We show that SKEL has more biomechanically accurate joint locations than SMPL, and the bones fit inside the body surface better than previous methods. By fitting SKEL to SMPL meshes we are able to "upgrade" existing human pose and shape datasets to include biomechanical parameters. SKEL provides a new tool to enable biomechanics in the wild, while also providing vision and graphics researchers with a better constrained and more realistic model of human articulation. The model, code, and data are available for research at https://skel.is.tue.mpg.de.

Given a series of natural language descriptions, our task is to generate 3D human motions that correspond semantically to the text, and follow the temporal order of the instructions. In particular, our goal is to enable the synthesis of a series of actions, which we refer to as temporal action composition. The current state of the art in text-conditioned motion synthesis only takes a single action or a single sentence as input. This is partially due to lack of suitable training data containing action sequences, but also due to the computational complexity of their non-autoregressive model formulation, which does not scale well to long sequences. In this work, we address both issues. First, we exploit the recent BABEL motion-text collection, which has a wide range of labeled actions, many of which occur in a sequence with transitions between them. Next, we design a Transformer-based approach that operates non-autoregressively within an action, but autoregressively within the sequence of actions. This hierarchical formulation proves effective in our experiments when compared with multiple baselines. Our approach, called TEACH for “TEmporal Action Compositions for Human motions ”, produces realistic human motions for a wide variety of actions and temporal compositions from language descriptions. To encourage work on this new task, we make our code available for research purposes at teach.is.tue.mpg.de.

Parametric 3D body models like SMPL only represent minimally-clothed people and are hard to extend to clothing because they have a fixed mesh topology and resolution. To address these limitations, recent work uses implicit surfaces or point clouds to model clothed bodies. While not limited by topology, such methods still struggle to model clothing that deviates significantly from the body, such as skirts and dresses. This is because they rely on the body to canonicalize the clothed surface by reposing it to a reference shape. Unfortunately, this process is poorly defined when clothing is far from the body. Additionally, they use linear blend skinning to pose the body and the skinning weights are tied to the underlying body parts. In contrast, we model the clothing deformation in a local coordinate space without canonicalization. We also relax the skinning weights to let multiple body parts influence the surface. Specifically, we extend point-based methods with a coarse stage, that replaces canonicalization with a learned pose-independent “coarse shape” that can capture the rough surface geometry of clothing like skirts. We then refine this using a network that infers the linear blend skinning weights and pose dependent displacements from the coarse representation. The approach works well for garments that both conform to, and deviate from, the body. We demonstrate the usefulness of our approach by learning person-specific avatars from examples and then show how they can be animated in new poses and motions. We also show that the method can learn directly from raw scans with missing data, greatly simplifying the process of creating realistic avatars. Code is available for research purposes at https://qianlim.github.io/SkiRT.

To make 3D human avatars widely available, we must be able to generate a variety of 3D virtual humans with varied identities and shapes in arbitrary poses. This task is chal-lenging due to the diversity of clothed body shapes, their complex articulations, and the resulting rich, yet stochas-tic geometric detail in clothing. Hence, current methods that represent 3D people do not provide a full generative model of people in clothing. In this paper, we propose a novel method that learns to generate detailed 3D shapes of people in a variety of garments with corresponding skin-ning weights. Specifically, we devise a multi-subject forward skinning module that is learned from only a few posed, unrigged scans per subject. To capture the stochastic nature of high-frequency details in garments, we leverage an adversarial loss formulation that encourages the model to capture the underlying statistics. We provide empirical evi-dence that this leads to realistic generation of local details such as wrinkles. We show that our model is able to gen-erate natural human avatars wearing diverse and detailed clothing. Furthermore, we show that our method can be used on the task of fitting human models to raw scans, out-performing the previous state-of-the-art.

While methods that regress 3D human meshes from images have progressed rapidly, the estimated body shapes often do not capture the true human shape. This is problematic since, for many applications, accurate body shape is as important as pose. The key reason that body shape accuracy lags pose accuracy is the lack of data. While humans can label 2D joints, and these constrain 3D pose, it is not so easy to “label” 3D body shape. Since paired data with images and 3D body shape are rare, we exploit two sources of information: (1) we collect internet images of diverse “fashion” models together with a small set of anthropometric measurements; (2) we collect linguistic shape attributes for a wide range of 3D body meshes and the model images. Taken together, these datasets provide sufficient constraints to infer dense 3D shape. We exploit the anthropometric measurements and linguistic shape attributes in several novel ways to train a neural network, called SHAPY, that regresses 3D human pose and shape from an RGB image. We evaluate SHAPY on public benchmarks, but note that they either lack significant body shape variation, ground-truth shape, or clothing variation. Thus, we collect a new dataset for evaluating 3D human shape estimation, called HBW, containing photos of “Human Bodies in the Wild” for which we have ground-truth 3D body scans. On this new benchmark, SHAPY significantly outperforms state-of-the-art methods on the task of 3D body shape estimation. This is the first demonstration that 3D body shape regression from images can be trained from easy-to-obtain anthropometric measurements and linguistic shape attributes. Our model and data are available at: shapy.is.tue.mpg.de

As 3D facial avatars become more widely used for communication, it is critical that they faithfully convey emotion. Unfortunately, the best recent methods that regress parametric 3D face models from monocular images are unable to capture the full spectrum of facial expression, such as subtle or extreme emotions. We find the standard reconstruction metrics used for training (landmark reprojection error, photometric error, and face recognition loss) are insufficient to capture high-fidelity expressions. The result is facial geometries that do not match the emotional content of the input image. We address this with EMOCA (EMOtion Capture and Animation), by introducing a novel deep perceptual emotion consistency loss during training, which helps ensure that the reconstructed 3D expression matches the expression depicted in the input image. While EMOCA achieves 3D reconstruction errors that are on par with the current best methods, it significantly outperforms them in terms of the quality of the reconstructed expression and the perceived emotional content. We also directly regress levels of valence and arousal and classify basic expressions from the estimated 3D face parameters. On the task of in-the-wild emotion recognition, our purely geometric approach is on par with the best image-based methods, highlighting the value of 3D geometry in analyzing human behavior. The model and code are publicly available at https://emoca.is.tue.mpg.de.

Inferring human-scene contact (HSC) is the first step toward understanding how humans interact with their surroundings. While detecting 2D human-object interaction (HOI) and reconstructing 3D human pose and shape (HPS) have enjoyed significant progress, reasoning about 3D human-scene contact from a single image is still challenging. Existing HSC detection methods consider only a few types of predefined contact, often reduce the body and scene to a small number of primitives, and even overlook image evidence. To predict human-scene contact from a single image, we address the limitations above from both data and algorithmic perspectives. We capture a new dataset called RICH for “Real scenes, Interaction, Contact and Humans.” RICH contains multiview outdoor/indoor video sequences at 4K resolution, ground-truth 3D human bodies captured using markerless motion capture, 3D body scans, and high resolution 3D scene scans. A key feature of RICH is that it also contains accurate vertex-level contact labels on the body. Using RICH, we train a network that predicts dense body-scene contacts from a single RGB image. Our key insight is that regions in contact are always occluded so the network needs the ability to explore the whole image for evidence. We use a transformer to learn such non-local relationships and propose a new Body-Scene contact TRansfOrmer (BSTRO). Very few methods explore 3D contact; those that do focus on the feet only, detect foot contact as a post-processing step, or infer contact from body pose without looking at the scene. To our knowledge, BSTRO is the first method to directly estimate 3D body-scene contact from a single image. We demonstrate that BSTRO significantly outperforms the prior art. Our code and dataset are available for research purposes at: https://rich.is.tue.mpg.de.

We address the problem of inferring the anatomic skeleton of a person, in an arbitrary pose, from the 3D surface of the body; i.e. we predict the inside (bones) from the outside (skin). This has many applications in medicine and biomechanics. Existing state-of-the-art biomechanical skeletons are detailed but do not easily generalize to new subjects. Additionally, computer vision and graphics methods that predict skeletons are typically heuristic, not learned from data, do not leverage the full 3D body surface, and are not validated against ground truth. To our knowledge, our system, called OSSO (Obtaining Skeletal Shape from Outside), is the first to learn the mapping from the 3D body surface to the internal skeleton from real data. We do so using 1000 male and 1000 female dual-energy X-ray absorptiometry (DXA) scans. To these, we fit a parametric 3D body shape model (STAR) to capture the body surface and a novel part-based 3D skeleton model to capture the bones. This provides inside/outside training pairs. We model the statistical variation of full skeletons using PCA in a pose-normalized space and train a regressor from body shape parameters to skeleton shape parameters. Given an arbitrary 3D body shape and pose, OSSO predicts a realistic skeleton inside. In contrast to previous work, we evaluate the accuracy of the skeleton shape quantitatively on held out DXA scans, outperforming the state-of-the art. We also show 3D skeleton prediction from varied and challenging 3D bodies. The code to infer a skeleton from a body shape is available at https://osso.is.tue.mpg.de, and the dataset of paired outer surface (skin) and skeleton (bone) meshes is available as a Biobank Returned Dataset. This research has been conducted using the UK Biobank Resource.

Our goal is to recover the 3D shape and pose of dogs from a single image. This is a challenging task because dogs exhibit a wide range of shapes and appearances, and are highly articulated. Recent work has proposed to directly regress the SMAL animal model, with additional limb scale parameters, from images. Our method, called BARC (Breed-Augmented Regression using Classification), goes beyond prior work in several important ways. First, we modify the SMAL shape space to be more appropriate for representing dog shape. But, even with a better shape model, the problem of regressing dog shape from an image is still challenging because we lack paired images with 3D ground truth. To compensate for the lack of paired data, we formulate novel losses that exploit information about dog breeds. In particular, we exploit the fact that dogs of the same breed have similar body shapes. We formulate a novel breed similarity loss consisting of two parts: One term encourages the shape of dogs from the same breed to be more similar than dogs of different breeds. The second one, a breed classification loss, helps to produce recognizable breed-specific shapes. Through ablation studies, we find that our breed losses significantly improve shape accuracy over a baseline without them. We also compare BARC qualitatively to WLDO with a perceptual study and find that our approach produces dogs that are significantly more realistic. This work shows that a-priori information about genetic similarity can help to compensate for the lack of 3D training data. This concept may be applicable to other animal species or groups of species. Our code is publicly available for research purposes at https://barc.is.tue.mpg.de/.

Given an image with multiple people, our goal is to directly regress the pose and shape of all the people as well as their relative depth. Inferring the depth of a person in an image, however, is fundamentally ambiguous without knowing their height. This is particularly problematic when the scene contains people of very different sizes, e.g. from infants to adults. To solve this, we need several things. First, we develop a novel method to infer the poses and depth of multiple people in a single image. While previous work that estimates multiple people does so by reasoning in the image plane, our method, called BEV, adds an additional imaginary Bird's-Eye-View representation to explicitly reason about depth. BEV reasons simultaneously about body centers in the image and in depth and, by combing these, estimates 3D body position. Unlike prior work, BEV is a single-shot method that is end-to-end differentiable. Second, height varies with age, making it impossible to resolve depth without also estimating the age of people in the image. To do so, we exploit a 3D body model space that lets BEV infer shapes from infants to adults. Third, to train BEV, we need a new dataset. Specifically, we create a “Relative Human” (RH) dataset that includes age labels and relative depth relationships between the people in the images. Extensive experiments on RH and AGORA demonstrate the effectiveness of the model and training scheme. BEV out-performs existing methods on depth reasoning, child shape estimation, and robustness to occlusion. The code11https://github.com/Arthur151/ROMP and dataset22https://github.com/Arthur151/Relative_Human are released for research purposes.

Generating digital humans that move realistically has many applications and is widely studied, but existing meth-odsfocus on the major limbs of the body, ignoring the hands and head. Hands have been separately studied, but the fo-cus has been on generating realistic static grasps of objects. To synthesize virtual characters that interact with the world, we need to generate full-body motions and realistic hand grasps simultaneously. Both sub-problems are challenging on their own and, together, the state space of poses is sig-nificantly larger, the scales of hand and body motions dif-fer, and the whole-body posture and the hand grasp must agree, satisfy physical constraints, and be plausible. Additionally, the head is involved because the avatar must look at the object to interact with it. For the first time, we ad-dress the problem of generating full-body, hand and head motions of an avatar grasping an unknown object. As in-put, our method, called GOAL, takes a 3D object, its pose, and a starting 3D body pose and shape. GOAL outputs a sequence of whole-body poses using two novel networks. First, GNet generates a goal whole-body grasp with a re-alistic body, head, arm, and hand pose, as well as hand-object contact. Second, MNet generates the motion be-tween the starting and goal pose. This is challenging, as it requires the avatar to walk towards the object with foot-ground contact, orient the head towards it, reach out, and grasp it with a realistic hand pose and hand-object con-tact. To achieve this the networks exploit a representation that combines SMPL-X body parameters and 3D vertex off-sets. We train and evaluate GOAL, both qualitatively and quantitatively, on the GRAB dataset. Results show that GOAL generalizes well to unseen objects, outperforming baselines. A perceptual study shows that GOAL's gener-ated motions approach the realism of GRAB's ground truth. GOAL takes a step towards generating realistic full-body object grasping motion. Our models and code are available at https://goal.is.tue.mpg.de.

Current methods for learning realistic and animatable 3D clothed avatars need either posed 3D scans or 2D images with carefully controlled user poses. In contrast, our goal is to learn an avatar from only 2D images of people in unconstrained poses. Given a set of images, our method estimates a detailed 3D surface from each image and then combines these into an animatable avatar. Implicit functions are well suited to the first task, as they can capture details like hair and clothes. Current methods, however, are not robust to varied human poses and often produce 3D surfaces with broken or disembodied limbs, missing details, or non-human shapes. The problem is that these methods use global feature encoders that are sensitive to global pose. To address this, we propose ICON (“Tmplicit Clothed humans Obtained from Normals”), which, instead, uses local features. ICON has two main modules, both of which exploit the SMPL(-X) body model. First, ICON infers detailed clothed-human normals (front/back) conditioned on the SMPL(-X) normals. Second, a visibility-aware implicit surface regressor produces an iso-surface of a human occupancy field. Importantly, at inference time, a feedback loop alternates between refining the SMPL(-X) mesh using the inferred clothed normals and then refining the normals. Given multiple reconstructed frames of a subject in varied poses, we use a modified version of SCANimate to produce an animatable avatar from them. Evaluation on the AGORA and CAPE datasets shows that ICON outperforms the state of the art in reconstruction, even with heavily limited training data. Additionally, it is much more robust to out-of-distribution samples, e.g., in-the-wild poses/images and out-of-frame cropping. ICON takes a step towards robust 3D clothed human reconstruction from in-the-wild images. This enables avatar creation directly from video with personalized pose-dependent cloth deformation. Models and code are available for research at https://icon.is.tue.mpg.de.

Humans are in constant contact with the world as they move through it and interact with it. This contact is a vital source of information for understanding 3D humans, 3D scenes, and the interactions between them. In fact, we demonstrate that these human-scene interactions (HSIs) can be leveraged to improve the 3D reconstruction of a scene from a monocular RGB video. Our key idea is that, as a person moves through a scene and interacts with it, we accumulate HSIs across multiple input images, and use these in optimizing the 3D scene to reconstruct a consistent, physically plausible, 3D scene layout. Our optimization-based approach exploits three types of HSI constraints: (1) humans who move in a scene are occluded by, or occlude, objects, thus constraining the depth ordering of the objects, (2) humans move throughfree space and do not interpenetrate objects, (3) when humans and objects are in contact, the contact surfaces occupy the same place in space. Using these constraints in an optimization formulation across all observations, we significantly improve 3D scene layout reconstruction. Furthermore, we show that our scene reconstruction can be used to refine the initial 3D human pose and shape (HPS) estimation. We evaluate the 3D scene layout reconstruction and HPS estimates qualitatively and quantitatively using the PROX and PiGraphs datasets. The code and data are available for research purposes at https://mover.is.tue.mpg.de.

Traditional 3D morphable face models (3DMMs) provide fine-grained control over expression but cannot easily capture geometric and appearance details. Neural volumetric representations approach photorealism but are hard to animate and do not generalize well to unseen expressions. To tackle this problem, we propose IMavatar (Implicit Morphable avatar), a novel method for learning implicit head avatars from monocular videos. Inspired by the fine-grained control mechanisms afforded by conventional 3DMMs, we represent the expression- and pose-related deformations via learned blendshapes and skinning fields. These attributes are pose-independent and can be used to morph the canonical geometry and texture fields given novel expression and pose parameters. We employ ray marching and iterative root-finding to locate the canonical surface intersection for each pixel. A key contribution is our novel analytical gradient formulation that enables end-to-end training of IMavatars from videos. We show quantitatively and qualitatively that our method improves geometry and covers a more complete expression space compared to state-of-the-art methods. Code and data can be found at https://ait.ethz.ch/projects/2022/IMavatar/.

Blimps are well suited to perform long-duration aerial tasks as they are energy efficient, relatively silent and safe. To address the blimp navigation and control task, in previous work we developed a hardware and software-in-the-loop framework and a PID-based controller for large blimps in the presence of wind disturbance. However, blimps have a deformable structure and their dynamics are inherently non-linear and time-delayed, making PID controllers difficult to tune. Thus, often resulting in large tracking errors. Moreover, the buoyancy of a blimp is constantly changing due to variations in ambient temperature and pressure. To address these issues, in this paper we present a learning-based framework based on deep residual reinforcement learning (DRRL), for the blimp control task. Within this framework, we first employ a PID controller to provide baseline performance. Subsequently, the DRRL agent learns to modify the PID decisions by interaction with the environment. We demonstrate in simulation that DRRL agent consistently improves the PID performance. Through rigorous simulation experiments, we show that the agent is robust to changes in wind speed and buoyancy. In real-world experiments, we demonstrate that the agent, trained only in simulation, is sufficiently robust to control an actual blimp in windy conditions. We openly provide the source code of our approach at https://github.com/robot-perception-group/AutonomousBlimpDRL. Video demonstration is provided at https://youtu.be/EMC4KnlH0yI.

While recent work has shown progress on extracting clothed 3D human avatars from a single image, video, or a set of 3D scans, several limitations remain. Most methods use a holistic representation to jointly model the body and clothing, which means that the clothing and body cannot be separated for applications like virtual try-on. Other methods separately model the body and clothing, but they require training from a large set of 3D clothed human meshes obtained from 3D/4D scanners or physics simulations. Our insight is that the body and clothing have different modeling requirements. While the body is well represented by a mesh-based parametric 3D model, implicit representations and neural radiance fields are better suited to capturing the large variety in shape and appearance present in clothing. Building on this insight, we propose SCARF (Segmented Clothed Avatar Radiance Field), a hybrid model combining a mesh-based body with a neural radiance field. Integrating the mesh into the volumetric rendering in combination with a differentiable rasterizer enables us to optimize SCARF directly from monocular videos, without any 3D supervision. The hybrid modeling enables SCARF to (i) animate the clothed body avatar by changing body poses (including hand articulation and facial expressions), (ii) synthesize novel views of the avatar, and (iii) transfer clothing between avatars in virtual try-on applications. We demonstrate that SCARF reconstructs clothing with higher visual quality than existing methods, that the clothing deforms with changing body pose and body shape, and that clothing can be successfully transferred between avatars of different subjects. The code and models are available at https://github.com/YadiraF/SCARF.

Occlusion boundaries contain rich perceptual information about the underlying scene structure and provide important cues in many visual perception-related tasks such as object recognition, segmentation, motion estimation, scene understanding, and autonomous navigation. However, there is no formal definition of occlusion boundaries in the literature, and state-of-the-art occlusion boundary detection is still suboptimal. With this in mind, in this paper we propose a formal definition of occlusion boundaries for related studies. Further, based on a novel idea, we develop two concrete approaches with different characteristics to detect occlusion boundaries in video sequences via enhanced exploration of contextual information (e.g, local structural boundary patterns, observations from surrounding regions, and temporal context) with deep models and conditional random fields. Experimental evaluations of our methods on two challenging occlusion boundary benchmarks (CMU and VSB100) demonstrate that our detectors significantly outperform the current state-of-the-art. Finally, we empirically assess the roles of several important components of the proposed detectors to validate the rationale behind these approaches.

In natural conversation and interaction, our hands often overlap or are in contact with each other. Due to the homogeneous appearance of hands, this makes estimating the 3D pose of interacting hands from images difficult. In this paper we demonstrate that self-similarity, and the resulting ambiguities in assigning pixel observations to the respective hands and their parts, is a major cause of the final 3D pose error. Motivated by this insight, we propose DIGIT, a novel method for estimating the 3D poses of two interacting hands from a single monocular image. The method consists of two interwoven branches that process the input imagery into a per-pixel semantic part segmentation mask and a visual feature volume. In contrast to prior work, we do not decouple the segmentation from the pose estimation stage, but rather leverage the per-pixel probabilities directly in the downstream pose estimation task. To do so, the part probabilities are merged with the visual features and processed via fully-convolutional layers. We experimentally show that the proposed approach achieves new state-of-the-art performance on the InterHand2.6M [33] dataset. We provide detailed ablation studies to demonstrate the efficacy of our method and to provide insights into how the modelling of pixel ownership affects 3D hand pose estimation.

Recovering expressive humans from images is essential for understanding human behavior. Methods that estimate 3D bodies, faces, or hands have progressed significantly, yet separately. Face methods recover accurate 3D shape and geometric details, but need a tight crop and struggle with extreme views and low resolution. Whole-body methods are robust to a wide range of poses and resolutions, but provide only a rough 3D face shape without details like wrinkles. To get the best of both worlds, we introduce PIXIE, which produces animatable, whole-body 3D avatars with realistic facial detail, from a single image. For this, PIXIE uses two key observations. First, existing work combines independent estimates from body, face, and hand experts, by trusting them equally. PIXIE introduces a novel moderator that merges the features of the experts, weighted by their confidence. All part experts can contribute to the whole, using SMPL-X’s shared shape space across all body parts. Second, human shape is highly correlated with gender, but existing work ignores this. We label training images as male, female, or non-binary, and train PIXIE to infer “gendered” 3D body shapes with a novel shape loss. In addition to 3D body pose and shape parameters, PIXIE estimates expression, illumination, albedo and 3D facial surface displacements. Quantitative and qualitative evaluation shows that PIXIE estimates more accurate whole-shape and detailed face shape than the state of the art. Models and code are available at pixie.is.tue.mpg.de.

Humans live within a 3D space and constantly interact with it to perform tasks. Such interactions involve physical contact between surfaces that is semantically meaningful. Our goal is to learn how humans interact with scenes and leverage this to enable virtual characters to do the same. To that end, we introduce a novel Human-Scene Interaction (HSI) model that encodes proximal relationships, called POSA for "Pose with prOximitieS and contActs". The representation of interaction is body-centric, which enables it to generalize to new scenes. Specifically, POSA augments the SMPL-X parametric human body model such that, for every mesh vertex, it encodes (a) the contact probability with the scene surface and (b) the corresponding semantic scene label. We learn POSA with a VAE conditioned on the SMPL-X vertices, and train on the PROX dataset, which contains SMPL-X meshes of people interacting with 3D scenes, and the corresponding scene semantics from the PROX-E dataset. We demonstrate the value of POSA with two applications. First, we automatically place 3D scans of people in scenes. We use a SMPL-X model fit to the scan as a proxy and then find its most likely placement in 3D. POSA provides an effective representation to search for "affordances" in the scene that match the likely contact relationships for that pose. We perform a perceptual study that shows significant improvement over the state of the art on this task. Second, we show that POSA’s learned representation of body-scene interaction supports monocular human pose estimation that is consistent with a 3D scene, improving on the state of the art. Our model and code are available for research purposes at https://posa.is.tue.mpg.de.

Learning to model and reconstruct humans in clothing is challenging due to articulation, non-rigid deformation, and varying clothing types and topologies. To enable learning, the choice of representation is the key. Recent work uses neural networks to parameterize local surface elements. This approach captures locally coherent geometry and non-planar details, can deal with varying topology, and does not require registered training data. However, naively using such methods to model 3D clothed humans fails to capture fine-grained local deformations and generalizes poorly. To address this, we present three key innovations: First, we deform surface elements based on a human body model such that large-scale deformations caused by articulation are explicitly separated from topological changes and local clothing deformations. Second, we address the limitations of existing neural surface elements by regressing local geometry from local features, significantly improving the expressiveness. Third, we learn a pose embedding on a 2D parameterization space that encodes posed body geometry, improving generalization to unseen poses by reducing non-local spurious correlations. We demonstrate the efficacy of our surface representation by learning models of complex clothing from point clouds. The clothing can change topology and deviate from the topology of the body. Once learned, we can animate previously unseen motions, producing high-quality point clouds, from which we generate realistic images with neural rendering. We assess the importance of each technical contribution and show that our approach outperforms the state-of-the-art methods in terms of reconstruction accuracy and inference time. The code is available for research purposes at https://qianlim.github.io/SCALE.

Substantial progress has been made on modeling rigid 3D objects using deep implicit representations. Yet, extending these methods to learn neural models of human shape is still in its infancy. Human bodies are complex and the key challenge is to learn a representation that generalizes such that it can express body shape deformations for unseen subjects in unseen, highly-articulated, poses. To address this challenge, we introduce LEAP (LEarning Articulated occupancy of People), a novel neural occupancy representation of the human body. Given a set of bone transformations (i.e. joint locations and rotations) and a query point in space, LEAP first maps the query point to a canonical space via learned linear blend skinning (LBS) functions and then efficiently queries the occupancy value via an occupancy network that models accurate identity- and pose-dependent deformations in the canonical space. Experiments show that our canonicalized occupancy estimation with the learned LBS functions greatly improves the generalization capability of the learned occupancy representation across various human shapes and poses, outperforming existing solutions in all settings.

While the accuracy of 3D human pose estimation from images has steadily improved on benchmark datasets, the best methods still fail in many real-world scenarios. This suggests that there is a domain gap between current datasets and common scenes containing people. To obtain ground-truth 3D pose, current datasets limit the complexity of clothing, environmental conditions, number of subjects, and occlusion. Moreover, current datasets evaluate sparse 3D joint locations corresponding to the major joints of the body, ignoring the hand pose and the face shape. To evaluate the current state-of-the-art methods on more challenging images, and to drive the field to address new problems, we introduce AGORA, a synthetic dataset with high realism and highly accurate ground truth. Here we use 4240 commercially-available, high-quality, textured human scans in diverse poses and natural clothing; this includes 257 scans of children. We create reference 3D poses and body shapes by fitting the SMPL-X body model (with face and hands) to the 3D scans, taking into account clothing. We create around 14K training and 3K test images by rendering between 5 and 15 people per image using either image-based lighting or rendered 3D environments, taking care to make the images physically plausible and photoreal. In total, AGORA consists of 173K individual person crops. We evaluate existing state-of-the-art methods for 3D human pose estimation on this dataset. and find that most methods perform poorly on images of children. Hence, we extend the SMPL-X model to better capture the shape of children. Additionally, we fine-tune methods on AGORA and show improved performance on both AGORA and 3DPW, confirming the realism of the dataset. We provide all the registered 3D reference training data, rendered images, and a web-based evaluation site at https://agora.is.tue.mpg.de/.

Understanding the semantics of human movement – the what, how and why of the movement – is an important problem that requires datasets of human actions with semantic labels. Existing datasets take one of two approaches. Large-scale video datasets contain many action labels but do not contain ground-truth 3D human motion. Alternatively, motion-capture (mocap) datasets have precise body motions but are limited to a small number of actions. To address this, we present BABEL, a large dataset with language labels describing the actions being performed in mocap sequences. BABEL consists of language labels for over 43 hours of mocap sequences from AMASS, containing over 250 unique actions. Each action label in BABEL is precisely aligned with the duration of the corresponding action in the mocap sequence. BABELalso allows overlap of multiple actions, that may each span different durations. This results in a total of over 66000 action segments. The dense annotations can be leveraged for tasks like action recognition, temporal localization, motion synthesis, etc. To demonstrate the value of BABEL as a benchmark, we evaluate the performance of models on 3D action recognition. We demonstrate that BABEL poses interesting learning challenges that are applicable to real-world scenarios, and can serve as a useful benchmark for progress in 3D action recognition. The dataset, baseline methods, and evaluation code are available and supported for academic research purposes at https://babel.is.tue.mpg.de/.

We present SCANimate, an end-to-end trainable framework that takes raw 3D scans of a clothed human and turns them into an animatable avatar. These avatars are driven by pose parameters and have realistic clothing that moves and deforms naturally. SCANimate does not rely on a customized mesh template or surface mesh registration. We observe that fitting a parametric 3D body model, like SMPL, to a clothed human scan is tractable while surface registration of the body topology to the scan is often not, because clothing can deviate significantly from the body shape. We also observe that articulated transformations are invertible, resulting in geometric cycle-consistency in the posed and unposed shapes. These observations lead us to a weakly supervised learning method that aligns scans into a canonical pose by disentangling articulated deformations without template-based surface registration. Furthermore, to complete missing regions in the aligned scans while modeling pose-dependent deformations, we introduce a locally pose-aware implicit function that learns to complete and model geometry with learned pose correctives. In contrast to commonly used global pose embeddings, our local pose conditioning significantly reduces long-range spurious correlations and improves generalization to unseen poses, especially when training data is limited. Our method can be applied to pose-aware appearance modeling to generate a fully textured avatar. We demonstrate our approach on various clothing types with different amounts of training data, outperforming existing solutions and other variants in terms of fidelity and generality in every setting. The code is available at https://scanimate.is.tue.mpg.de.

A key step towards understanding human behavior is the prediction of 3D human motion. Successful solutions have many applications in human tracking, HCI, and graphics. Most previous work focuses on predicting a time series of future 3D joint locations given a sequence 3D joints from the past. This Euclidean formulation generally works better than predicting pose in terms of joint rotations. Body joint locations, however, do not fully constrain 3D human pose, leaving degrees of freedom (like rotation about a limb) undefined. Note that 3D joints can be viewed as a sparse point cloud. Thus the problem of human motion prediction can be seen as a problem of point cloud prediction. With this observation, we instead predict a sparse set of locations on the body surface that correspond to motion capture markers. Given such markers, we fit a parametric body model to recover the 3D body of the person. These sparse surface markers also carry detailed information about human movement that is not present in the joints, increasing the naturalness of the predicted motions. Using the AMASS dataset, we train MOJO (More than Our JOints), which is a novel variational autoencoder with a latent DCT space that generates motions from latent frequencies. MOJO preserves the full temporal resolution of the input motion, and sampling from the latent frequencies explicitly introduces high-frequency components into the generated motion. We note that motion prediction methods accumulate errors over time, resulting in joints or markers that diverge from true human bodies. To address this, we fit the SMPL-X body model to the predictions at each time step, projecting the solution back onto the space of valid bodies, before propagating the new markers in time. Quantitative and qualitative experiments show that our approach produces state-of-the-art results and realistic 3D body animations. The code is available for research purposes at https://yz-cnsdqz.github.io/MOJO/MOJO.html.

Neural implicit surface representations have emerged as a promising paradigm to capture 3D shapes in a continuous and resolution-independent manner. However, adapting them to articulated shapes is non-trivial. Existing approaches learn a backward warp field that maps deformed to canonical points. However, this is problematic since the backward warp field is pose dependent and thus requires large amounts of data to learn. To address this, we introduce SNARF, which combines the advantages of linear blend skinning (LBS) for polygonal meshes with those of neural implicit surfaces by learning a forward deformation field without direct supervision. This deformation field is defined in canonical, pose-independent, space, enabling generalization to unseen poses. Learning the deformation field from posed meshes alone is challenging since the correspondences of deformed points are defined implicitly and may not be unique under changes of topology. We propose a forward skinning model that finds all canonical correspondences of any deformed point using iterative root finding. We derive analytical gradients via implicit differentiation, enabling end-to-end training from 3D meshes with bone transformations. Compared to state-of-the-art neural implicit representations, our approach generalizes better to unseen poses while preserving accuracy. We demonstrate our method in challenging scenarios on (clothed) 3D humans in diverse and unseen poses.

Learning to regress 3D human body shape and pose (e.g. SMPL parameters) from monocular images typically exploits losses on 2D keypoints, silhouettes, and/or part-segmentation when 3D training data is not available. Such losses, however, are limited because 2D keypoints do not supervise body shape and segmentations of people in clothing do not match projected minimally-clothed SMPL shapes. To exploit richer image information about clothed people, we introduce higher-level semantic information about clothing to penalize clothed and non-clothed regions of the human body differently. To do so, we train a body regressor using a novel "Differentiable Semantic Rendering (DSR)" loss. For Minimally-Clothed (MC) regions, we define the DSR-MC loss, which encourages a tight match between a rendered SMPL body and the minimally-clothed regions of the image. For clothed regions, we define the DSR-C loss to encourage the rendered SMPL body to be inside the clothing mask. To ensure end-to-end differentiable training, we learn a semantic clothing prior for SMPL vertices from thousands of clothed human scans. We perform extensive qualitative and quantitative experiments to evaluate the role of clothing semantics on the accuracy of 3D human pose and shape estimation. We outperform all previous state-of-the-art methods on 3DPW and Human3.6M and obtain on par results on MPI-INF-3DHP. Code and trained models are available for research at https://dsr.is.tue.mpg.de/.

Marker-based optical motion capture (mocap) is the "gold standard" method for acquiring accurate 3D human motion in computer vision, medicine, and graphics. The raw output of these systems are noisy and incomplete 3D points or short tracklets of points. To be useful, one must associate these points with corresponding markers on the captured subject; i.e. "labelling". Given these labels, one can then "solve" for the 3D skeleton or body surface mesh. Commercial auto-labeling tools require a specific calibration procedure at capture time, which is not possible for archival data. Here we train a novel neural network called SOMA, which takes raw mocap point clouds with varying numbers of points, labels them at scale without any calibration data, independent of the capture technology, and requiring only minimal human intervention. Our key insight is that, while labeling point clouds is highly ambiguous, the 3D body provides strong constraints on the solution that can be exploited by a learning-based method. To enable learning, we generate massive training sets of simulated noisy and ground truth mocap markers animated by 3D bodies from AMASS. SOMA exploits an architecture with stacked self-attention elements to learn the spatial structure of the 3D body and an optimal transport layer to constrain the assignment (labeling) problem while rejecting outliers. We extensively evaluate SOMA both quantitatively and qualitatively. SOMA is more accurate and robust than existing state of the art research methods and can be applied where commercial systems cannot. We automatically label over 8 hours of archival mocap data across 4 different datasets captured using various technologies and output SMPL-X body models. The model and data is released for research purposes at https://soma.is.tue.mpg.de/.

A long-standing goal in computer vision is to capture, model, and realistically synthesize human behavior. Specifically, by learning from data, our goal is to enable virtual humans to navigate within cluttered indoor scenes and naturally interact with objects. Such embodied behavior has applications in virtual reality, computer games, and robotics, while synthesized behavior can be used as training data. The problem is challenging because real human motion is diverse and adapts to the scene. For example, a person can sit or lie on a sofa in many places and with varying styles. We must model this diversity to synthesize virtual humans that realistically perform human-scene interactions. We present a novel data-driven, stochastic motion synthesis method that models different styles of performing a given action with a target object. Our Scene-Aware Motion Prediction method (SAMP) generalizes to target objects of various geometries while enabling the character to navigate in cluttered scenes. To train SAMP, we collected MoCap data covering various sitting, lying down, walking, and running styles. We demonstrate SAMP on complex indoor scenes and achieve superior performance than existing solutions. Code and data are available for research at https://samp.is.tue.mpg.de.

Despite significant progress, we show that state of the art 3D human pose and shape estimation methods remain sensitive to partial occlusion and can produce dramatically wrong predictions although much of the body is observable. To address this, we introduce a soft attention mechanism, called the Part Attention REgressor (PARE), that learns to predict body-part-guided attention masks. We observe that state-of-the-art methods rely on global feature representations, making them sensitive to even small occlusions. In contrast, PARE’s part-guided attention mechanism overcomes these issues by exploiting information about the visibility of individual body parts while leveraging information from neighboring body-parts to predict occluded parts. We show qualitatively that PARE learns sensible attention masks, and quantitative evaluation confirms that PARE achieves more accurate and robust reconstruction results than existing approaches on both occlusion-specific and standard benchmarks. The code and data are available for research purposes at https://pare.is.tue.mpg.de/

Due to the lack of camera parameter information for in-the-wild images, existing 3D human pose and shape (HPS) estimation methods make several simplifying assumptions: weak-perspective projection, large constant focal length, and zero camera rotation. These assumptions often do not hold and we show, quantitatively and qualitatively, that they cause errors in the reconstructed 3D shape and pose. To address this, we introduce SPEC, the first in-the-wild 3D HPS method that estimates the perspective camera from a single image and employs this to reconstruct 3D human bodies more accurately. First, we train a neural network to estimate the field of view, camera pitch, and roll given an input image. We employ novel losses that improve the calibration accuracy over previous work. We then train a novel network that concatenates the camera calibration to the image features and uses these together to regress 3D body shape and pose. SPEC is more accurate than the prior art on the standard benchmark (3DPW) as well as two new datasets with more challenging camera views and varying focal lengths. Specifically, we create a new photorealistic synthetic dataset (SPEC-SYN) with ground truth 3D bodies and a novel in-the-wild dataset (SPEC-MTP) with calibration and high-quality reference bodies. Code and datasets are available for research purposes at https://spec.is.tue.mpg.de/.

Currently it requires an artist to create 3D human avatars with realistic clothing that can move naturally. Despite progress on 3D scanning and modeling of human bodies, there is still no technology that can easily turn a static scan into an animatable avatar. Automating the creation of such avatars would enable many applications in games, social networking, animation, and AR/VR to name a few. The key problem is one of representation. Standard 3D meshes are widely used in modeling the minimally-clothed body but do not readily capture the complex topology of clothing. Recent interest has shifted to implicit surface models for this task but they are computationally heavy and lack compatibility with existing 3D tools. What is needed is a 3D representation that can capture varied topology at high resolution and that can be learned from data. We argue that this representation has been with us all along — the point cloud. Point clouds have properties of both implicit and explicit representations that we exploit to model 3D garment geometry on a human body. We train a neural network with a novel local clothing geometric feature to represent the shape of different outfits. The network is trained from 3D point clouds of many types of clothing, on many bodies, in many poses, and learns to model pose-dependent clothing deformations. The geometry feature can be optimized to fit a previously unseen scan of a person in clothing, enabling the scan to be reposed realistically. Our model demonstrates superior quantitative and qualitative results in both multi-outfit modeling and unseen outfit animation. The code is available for research purposes at https://qianlim.github.io/POP.

We tackle the problem of action-conditioned generation of realistic and diverse human motion sequences. In contrast to methods that complete, or extend, motion sequences, this task does not require an initial pose or sequence. Here we learn an action-aware latent representation for human motions by training a generative variational autoencoder (VAE). By sampling from this latent space and querying a certain duration through a series of positional encodings, we synthesize variable-length motion sequences conditioned on a categorical action. Specifically, we design a Transformer-based architecture, ACTOR, for encoding and decoding a sequence of parametric SMPL human body models estimated from action recognition datasets. We evaluate our approach on the NTU RGB+D, HumanAct12 and UESTC datasets and show improvements over the state of the art. Furthermore, we present two use cases: improving action recognition through adding our synthesized data to training, and motion denoising. Code and models are available on our project page [53].

Synthesizing images of a person in novel poses from a single image is a highly ambiguous task. Most existing approaches require paired training images; i.e. images of the same person with the same clothing in different poses. However, obtaining sufficiently large datasets with paired data is challenging and costly. Previous methods that forego paired supervision lack realism. We propose a self-supervised framework named SPICE (Self-supervised Person Image CrEation) that closes the image quality gap with supervised methods. The key insight enabling self-supervision is to exploit 3D information about the human body in several ways. First, the 3D body shape must remain unchanged when reposing. Second, representing body pose in 3D enables reasoning about self occlusions. Third, 3D body parts that are visible before and after reposing, should have similar appearance features. Once trained, SPICE takes an image of a person and generates a new image of that person in a new target pose. SPICE achieves state-of-the-art performance on the DeepFashion dataset, improving the FID score from 29.9 to 7.8 compared with previous unsupervised methods, and with performance similar to the state-of-the-art supervised method (6.4). SPICE also generates temporally coherent videos given an input image and a sequence of poses, despite being trained on static images only.

This paper focuses on the regression of multiple 3D people from a single RGB image. Existing approaches predominantly follow a multi-stage pipeline that first detects people in bounding boxes and then independently regresses their 3D body meshes. In contrast, we propose to Regress all meshes in a One-stage fashion for Multiple 3D People (termed ROMP). The approach is conceptually simple, bounding box-free, and able to learn a per-pixel representation in an end-to-end manner. Our method simultaneously predicts a Body Center heatmap and a Mesh Parameter map, which can jointly describe the 3D body mesh on the pixel level. Through a body-center-guided sampling process, the body mesh parameters of all people in the image are easily extracted from the Mesh Parameter map. Equipped with such a fine-grained representation, our one-stage framework is free of the complex multi-stage process and more robust to occlusion. Compared with state-of-the-art methods, ROMP achieves superior performance on the challenging multi-person benchmarks, including 3DPW and CMU Panoptic. Experiments on crowded/occluded datasets demonstrate the robustness under various types of occlusion. The code, released at https://github.com/Arthur151/ROMP, is the first real-time implementation of monocular multi-person 3D mesh regression.

Recent advances in deep generative models have led to an unprecedented level of realism for synthetically generated images of humans. However, one of the remaining fundamental limitations of these models is the ability to flexibly control the generative process, e.g. change the camera and human pose while retaining the subject identity. At the same time, deformable human body models like SMPL [34] and its successors provide full control over pose and shape, but rely on classic computer graphics pipelines for rendering. Such rendering pipelines require explicit mesh rasterization that (a) does not have the potential to fix artifacts or lack of realism in the original 3D geometry and (b) until recently, were not fully incorporated into deep learning frameworks. In this work, we propose to bridge the gap between classic geometry-based rendering and the latest generative networks operating in pixel space. We train a network that directly converts a sparse set of 3D mesh vertices into photorealistic images, alleviating the need for traditional rasterization mechanism. We train our model on a large corpus of human 3D models and corresponding real photos, and show the advantage over conventional differentiable renderers both in terms of the level of photorealism and rendering efficiency.

While current monocular 3D face reconstruction methods can recover fine geometric details, they suffer several limitations. Some methods produce faces that cannot be realistically animated because they do not model how wrinkles vary with expression. Other methods are trained on high-quality face scans and do not generalize well to in-the-wild images. We present the first approach that regresses 3D face shape and animatable details that are specific to an individual but change with expression. Our model, DECA (Detailed Expression Capture and Animation), is trained to robustly produce a UV displacement map from a low-dimensional latent representation that consists of person-specific detail parameters and generic expression parameters, while a regressor is trained to predict detail, shape, albedo, expression, pose and illumination parameters from a single image. To enable this, we introduce a novel detail-consistency loss that disentangles person-specific details from expression-dependent wrinkles. This disentanglement allows us to synthesize realistic person-specific wrinkles by controlling expression parameters while keeping person-specific details unchanged. DECA is learned from in-the-wild images with no paired 3D supervision and achieves state-of-the-art shape reconstruction accuracy on two benchmarks. Qualitative results on in-the-wild data demonstrate DECA's robustness and its ability to disentangle identity- and expression-dependent details enabling animation of reconstructed faces. The model and code are publicly available at https://deca.is.tue.mpg.de.

Photo-realistic visualization and animation of expressive human faces have been a long standing challenge. 3D face modeling methods provide parametric control but generates unrealistic images, on the other hand, generative 2D models like GANs (Generative Adversarial Networks) output photo-realistic face images, but lack explicit control. Recent methods gain partial control, either by attempting to disentangle different factors in an unsupervised manner, or by adding control post hoc to a pre-trained model. Unconditional GANs, however, may entangle factors that are hard to undo later. We condition our generative model on pre-defined control parameters to encourage disentanglement in the generation process. Specifically, we condition StyleGAN2 on FLAME, a generative 3D face model. While conditioning on FLAME parameters yields unsatisfactory results, we find that conditioning on rendered FLAME geometry and photometric details works well. This gives us a generative 2D face model named GIF (Generative Interpretable Faces) that offers FLAME's parametric control. Here, interpretable refers to the semantic meaning of different parameters. Given FLAME parameters for shape, pose, expressions, parameters for appearance, lighting, and an additional style vector, GIF outputs photo-realistic face images. We perform an AMT based perceptual study to quantitatively and qualitatively evaluate how well GIF follows its conditioning. The code, data, and trained model are publicly available for research purposes at http://gif.is.tue.mpg.de.

Robotic grasping of house-hold objects has made remarkable progress in recent years. Yet, human grasps are still difficult to synthesize realistically. There are several key reasons: (1) the human hand has many degrees of freedom (more than robotic manipulators); (2) the synthesized hand should conform to the surface of the object; and (3) it should interact with the object in a semantically and physically plausible manner. To make progress in this direction, we draw inspiration from the recent progress on learning-based implicit representations for 3D object reconstruction. Specifically, we propose an expressive representation for human grasp modelling that is efficient and easy to integrate with deep neural networks. Our insight is that every point in a three-dimensional space can be characterized by the signed distances to the surface of the hand and the object, respectively. Consequently, the hand, the object, and the contact area can be represented by implicit surfaces in a common space, in which the proximity between the hand and the object can be modelled explicitly. We name this 3D to 2D mapping as Grasping Field, parameterize it with a deep neural network, and learn it from data. We demonstrate that the proposed grasping field is an effective and expressive representation for human grasp generation. Specifically, our generative model is able to synthesize high-quality human grasps, given only on a 3D object point cloud. The extensive experiments demonstrate that our generative model compares favorably with a strong baseline and approaches the level of natural human grasps. Furthermore, based on the grasping field representation, we propose a deep network for the challenging task of 3D hand-object interaction reconstruction from a single RGB image. Our method improves the physical plausibility of the hand-object contact reconstruction and achieves comparable performance for 3D hand reconstruction compared to state-of-the-art methods. Our model and code are available for research purpose at https://github.com/korrawe/grasping_field.

The following topics are dealt with: image reconstruction; learning (artificial intelligence); solid modelling; image representation; feature extraction; neural nets; stereo image processing; image colour analysis; cameras; convolutional neural nets.

The goal of many computer vision systems is to transform image pixels into 3D representations. Recent popular models use neural networks to regress directly from pixels to 3D object parameters. Such an approach works well when supervision is available, but in problems like human pose and shape estimation, it is difficult to obtain natural images with 3D ground truth. To go one step further, we propose a new architecture that facilitates unsupervised, or lightly supervised, learning. The idea is to break the problem into a series of transformations between increasingly abstract representations. Each step involves a cycle designed to be learnable without annotated training data, and the chain of cycles delivers the final solution. Specifically, we use 2D body part segments as an intermediate representation that contains enough information to be lifted to 3D, and at the same time is simple enough to be learned in an unsupervised way. We demonstrate the method by learning 3D human pose and shape from un-paired and un-annotated images. We also explore varying amounts of paired data and show that cycling greatly alleviates the need for paired data. While we present results for modeling humans, our formulation is general and can be applied to other vision problems.

We present a fully automatic system that takes a 3D scene and generates plausible 3D human bodies that are posed naturally in that 3D scene. Given a 3D scene without people, humans can easily imagine how people could interact with the scene and the objects in it. However, this is a challenging task for a computer as solving it requires that (1) the generated human bodies to be semantically plausible within the 3D environment (e.g. people sitting on the sofa or cooking near the stove), and (2) the generated human-scene interaction to be physically feasible such that the human body and scene do not interpenetrate while, at the same time, body-scene contact supports physical interactions. To that end, we make use of the surface-based 3D human model SMPL-X. We first train a conditional variational autoencoder to predict semantically plausible 3D human poses conditioned on latent scene representations, then we further refine the generated 3D bodies using scene constraints to enforce feasible physical interaction. We show that our approach is able to synthesize realistic and expressive 3D human bodies that naturally interact with 3D environment. We perform extensive experiments demonstrating that our generative framework compares favorably with existing methods, both qualitatively and quantitatively. We believe that our scene-conditioned 3D human generation pipeline will be useful for numerous applications; e.g. to generate training data for human pose estimation, in video games and in VR/AR. Our project page for data and code can be seen at: {https://vlg.inf.ethz.ch/projects/PSI/}.

Human motion is fundamental to understanding behavior. Despite progress on single-image 3D pose and shape estimation, existing video-based state-of-the-art methods fail to produce accurate and natural motion sequences due to a lack of ground-truth 3D motion data for training. To address this problem, we propose "Video Inference for Body Pose and Shape Estimation'' (VIBE), which makes use of an existing large-scale motion capture dataset (AMASS) together with unpaired, in-the-wild, 2D keypoint annotations. Our key novelty is an adversarial learning framework that leverages AMASS to discriminate between real human motions and those produced by our temporal pose and shape regression networks. We define a novel temporal network architecture with a self-attention mechanism and show that adversarial training, at the sequence level, produces kinematically plausible motion sequences without in-the-wild ground-truth 3D labels. We perform extensive experimentation to analyze the importance of motion and demonstrate the effectiveness of VIBE on challenging 3D pose estimation datasets, achieving state-of-the-art performance. Code and pretrained models are available at https://github.com/mkocabas/VIBE

Three-dimensional human body models are widely used in the analysis of human pose and motion. Existing models, however, are learned from minimally-clothed 3D scans and thus do not generalize to the complexity of dressed people in common images and videos. Additionally, current models lack the expressive power needed to represent the complex non-linear geometry of pose-dependent clothing shapes. To address this, we learn a generative 3D mesh model of clothed people from 3D scans with varying pose and clothing. Specifically, we train a conditional Mesh-VAE-GAN to learn the clothing deformation from the SMPL body model, making clothing an additional term in SMPL. Our model is conditioned on both pose and clothing type, giving the ability to draw samples of clothing to dress different body shapes in a variety of styles and poses. To preserve wrinkle detail, our Mesh-VAE-GAN extends patchwise discriminators to 3D meshes. Our model, named CAPE, represents global shape and fine local structure, effectively extending the SMPL body model to clothing. To our knowledge, this is the first generative model that directly dresses 3D human body meshes and generalizes to different poses. The model, code and data are available for research purposes at https://cape.is.tue.mpg.de.

Abstract The optical flow of humans is well known to be useful for the analysis of human action. Recent optical flow methods focus on training deep networks to approach the problem. However, the training data used by them does not cover the domain of human motion. Therefore, we develop a dataset of multi-human optical flow and train optical flow networks on this dataset. We use a 3D model of the human body and motion capture data to synthesize realistic flow fields in both single- and multi-person images. We then train optical flow networks to estimate human flow fields from pairs of images. We demonstrate that our trained networks are more accurate than a wide range of top methods on held-out test data and that they can generalize well to real image sequences. The code, trained models and the dataset are available for research.

Statistical models of the human body surface are generally learned from thousands of high-quality 3D scans in predefined poses to cover the wide variety of human body shapes and articulations. Acquisition of such data requires expensive equipment, calibration procedures, and is limited to cooperative subjects who can understand and follow instructions, such as adults. We present a method for learning a statistical 3D Skinned Multi-Infant Linear body model (SMIL) from incomplete, low-quality RGB-D sequences of freely moving infants. Quantitative experiments show that SMIL faithfully represents the RGB-D data and properly factorizes the shape and pose of the infants. To demonstrate the applicability of SMIL, we fit the model to RGB-D sequences of freely moving infants and show, with a case study, that our method captures enough motion detail for General Movements Assessment (GMA), a method used in clinical practice for early detection of neurodevelopmental disorders in infants. SMIL provides a new tool for analyzing infant shape and movement and is a step towards an automated system for GMA.

Susceptibility of deep neural networks to adversarial attacks poses a major theoretical and practical challenge. All efforts to harden classifiers against such attacks have seen limited success till now. Two distinct categories of samples against which deep neural networks are vulnerable, “adversarial samples” and “fooling samples”, have been tackled separately so far due to the difficulty posed when considered together. In this work, we show how one can defend against them both under a unified framework. Our model has the form of a variational autoencoder with a Gaussian mixture prior on the latent variable, such that each mixture component corresponds to a single class. We show how selective classification can be performed using this model, thereby causing the adversarial objective to entail a conflict. The proposed method leads to the rejection of adversarial samples instead of misclassification, while maintaining high precision and recall on test data. It also inherently provides a way of learning a selective classifier in a semi-supervised scenario, which can similarly resist adversarial attacks. We further show how one can reclassify the detected adversarial samples by iterative optimization.1

The creation of realistic self-avatars that users identify with is important for many virtual reality applications. However, current approaches for creating biometrically plausible avatars that represent a particular individual require expertise and are time-consuming. We investigated the visual perception of an avatar’s body dimensions by asking males and females to estimate their own body weight and shape on a virtual body using a virtual reality avatar creation tool. In a method of adjustment task, the virtual body was presented in an HTC Vive head-mounted display either co-located with (first-person perspective) or facing (third-person perspective) the participants. Participants adjusted the body weight and dimensions of various body parts to match their own body shape and size. Both males and females underestimated their weight by 10-20% in the virtual body, but the estimates of the other body dimensions were relatively accurate and within a range of ± 6%. There was a stronger influence of visual perspective on the estimates for males, but this effect was dependent on the amount of control over the shape of the virtual body, indicating that the results might be caused by where in the body the weight changes expressed themselves. These results suggest that this avatar creation tool could be used to allow participants to make a relatively accurate self-avatar in terms of adjusting body part dimensions, but not weight, and that the influence of visual perspective and amount of control needed over the body shape are likely gender-specific.

Audio-driven 3D facial animation has been widely explored, but achieving realistic, human-like performance is still unsolved. This is due to the lack of available 3D datasets, models, and standard evaluation metrics. To address this, we introduce a unique 4D face dataset with about 29 minutes of 4D scans captured at 60 fps and synchronized audio from 12 speakers. We then train a neural network on our dataset that factors identity from facial motion. The learned model, VOCA (Voice Operated Character Animation) takes any speech signal as input—even speech in languages other than English—and realistically animates a wide range of adult faces. Conditioning on subject labels during training allows the model to learn a variety of realistic speaking styles. VOCA also provides animator controls to alter speaking style, identity-dependent facial shape, and pose (i.e. head, jaw, and eyeball rotations) during animation. To our knowledge, VOCA is the only realistic 3D facial animation model that is readily applicable to unseen subjects without retargeting. This makes VOCA suitable for tasks like in-game video, virtual reality avatars, or any scenario in which the speaker, speech, or language is not known in advance. We make the dataset and model available for research purposes at http://voca.is.tue.mpg.de.

Estimating hand-object manipulations is essential for in- terpreting and imitating human actions. Previous work has made significant progress towards reconstruction of hand poses and object shapes in isolation. Yet, reconstructing hands and objects during manipulation is a more challeng- ing task due to significant occlusions of both the hand and object. While presenting challenges, manipulations may also simplify the problem since the physics of contact re- stricts the space of valid hand-object configurations. For example, during manipulation, the hand and object should be in contact but not interpenetrate. In this work, we regu- larize the joint reconstruction of hands and objects with ma- nipulation constraints. We present an end-to-end learnable model that exploits a novel contact loss that favors phys- ically plausible hand-object constellations. Our approach improves grasp quality metrics over baselines, using RGB images as input. To train and evaluate the model, we also propose a new large-scale synthetic dataset, ObMan, with hand-object manipulations. We demonstrate the transfer- ability of ObMan-trained models to real data.

To facilitate the analysis of human actions, interactions and emotions, we compute a 3D model of human body pose, hand pose, and facial expression from a single monocular image. To achieve this, we use thousands of 3D scans to train a new, unified, 3D model of the human body, SMPL-X, that extends SMPL with fully articulated hands and an expressive face. Learning to regress the parameters of SMPL-X directly from images is challenging without paired images and 3D ground truth. Consequently, we follow the approach of SMPLify, which estimates 2D features and then optimizes model parameters to fit the features. We improve on SMPLify in several significant ways: (1) we detect 2D features corresponding to the face, hands, and feet and fit the full SMPL-X model to these; (2) we train a new neural network pose prior using a large MoCap dataset; (3) we define a new interpenetration penalty that is both fast and accurate; (4) we automatically detect gender and the appropriate body models (male, female, or neutral); (5) our PyTorch implementation achieves a speedup of more than 8x over Chumpy. We use the new method, SMPLify-X, to fit SMPL-X to both controlled images and images in the wild. We evaluate 3D accuracy on a new curated dataset comprising 100 images with pseudo ground-truth. This is a step towards automatic expressive human capture from monocular RGB data. The models, code, and data are available for research purposes at https://smpl-x.is.tue.mpg.de.

We address the unsupervised learning of several interconnected problems in low-level vision: single view depth prediction, camera motion estimation, optical flow, and segmentation of a video into the static scene and moving regions. Our key insight is that these four fundamental vision problems are coupled through geometric constraints. Consequently, learning to solve them together simplifies the problem because the solutions can reinforce each other. We go beyond previous work by exploiting geometry more explicitly and segmenting the scene into static and moving regions. To that end, we introduce Competitive Collaboration, a framework that facilitates the coordinated training of multiple specialized neural networks to solve complex problems. Competitive Collaboration works much like expectation-maximization, but with neural networks that act as both competitors to explain pixels that correspond to static or moving regions, and as collaborators through a moderator that assigns pixels to be either static or independently moving. Our novel method integrates all these problems in a common framework and simultaneously reasons about the segmentation of the scene into moving objects and the static background, the camera motion, depth of the static scene structure, and the optical flow of moving objects. Our model is trained without any supervision and achieves state-of-the-art performance among joint unsupervised methods on all sub-problems.

The estimation of 3D face shape from a single image must be robust to variations in lighting, head pose, expression, facial hair, makeup, and occlusions. Robustness requires a large training set of in-the-wild images, which by construction, lack ground truth 3D shape. To train a network without any 2D-to-3D supervision, we present RingNet, which learns to compute 3D face shape from a single image. Our key observation is that an individual’s face shape is constant across images, regardless of expression, pose, lighting, etc. RingNet leverages multiple images of a person and automatically detected 2D face features. It uses a novel loss that encourages the face shape to be similar when the identity is the same and different for different people. We achieve invariance to expression by representing the face using the FLAME model. Once trained, our method takes a single image and outputs the parameters of FLAME, which can be readily animated. Additionally we create a new database of faces “not quite in-the-wild” (NoW) with 3D head scans and high-resolution images of the subjects in a wide variety of conditions. We evaluate publicly available methods and find that RingNet is more accurate than methods that use 3D supervision. The dataset, model, and results are available for research purposes at http://ringnet.is.tuebingen.mpg.de.

Deep neural networks provide powerful tools for pattern recognition, while classical graph algorithms are widely used to solve combinatorial problems. In computer vision, many tasks combine elements of both pattern recognition and graph reasoning. In this paper, we study how to connect deep networks with graph decomposition into an end-to-end trainable framework. More specifically, the minimum cost multicut problem is first converted to an unconstrained binary cubic formulation where cycle consistency constraints are incorporated into the objective function. The new optimization problem can be viewed as a Conditional Random Field (CRF) in which the random variables are associated with the binary edge labels. Cycle constraints are introduced into the CRF as high-order potentials. A standard Convolutional Neural Network (CNN) provides the front-end features for the fully differentiable CRF. The parameters of both parts are optimized in an end-to-end manner. The efficacy of the proposed learning algorithm is demonstrated via experiments on clustering MNIST images and on the challenging task of real-world multi-people pose estimation.

To understand and analyze human behavior, we need to capture humans moving in, and interacting with, the world. Most existing methods perform 3D human pose estimation without explicitly considering the scene. We observe however that the world constrains the body and vice-versa. To motivate this, we show that current 3D human pose estimation methods produce results that are not consistent with the 3D scene. Our key contribution is to exploit static 3D scene structure to better estimate human pose from monocular images. The method enforces Proximal Relationships with Object eXclusion and is called PROX. To test this, we collect a new dataset composed of 12 different 3D scenes and RGB sequences of 20 subjects moving in and interacting with the scenes. We represent human pose using the 3D human body model SMPL-X and extend SMPLify-X to estimate body pose using scene constraints. We make use of the 3D scene information by formulating two main constraints. The inter-penetration constraint penalizes intersection between the body model and the surrounding 3D scene. The contact constraint encourages specific parts of the body to be in contact with scene surfaces if they are close enough in distance and orientation. For quantitative evaluation we capture a separate dataset with 180 RGB frames in which the ground-truth body pose is estimated using a motion capture system. We show quantitatively that introducing scene constraints significantly reduces 3D joint error and vertex error. Our code and data are available for research at https://prox.is.tue.mpg.de.

Model-based human pose estimation is currently approached through two different paradigms. Optimization-based methods fit a parametric body model to 2D observations in an iterative manner, leading to accurate image-model alignments, but are often slow and sensitive to the initialization. In contrast, regression-based methods, that use a deep network to directly estimate the model parameters from pixels, tend to provide reasonable, but not pixel accurate, results while requiring huge amounts of supervision. In this work, instead of investigating which approach is better, our key insight is that the two paradigms can form a strong collaboration. A reasonable, directly regressed estimate from the network can initialize the iterative optimization making the fitting faster and more accurate. Similarly, a pixel accurate fit from iterative optimization can act as strong supervision for the network. This is the core of our proposed approach SPIN (SMPL oPtimization IN the loop). The deep network initializes an iterative optimization routine that fits the body model to 2D joints within the training loop, and the fitted estimate is subsequently used to supervise the network. Our approach is self-improving by nature, since better network estimates can lead the optimization to better solutions, while more accurate optimization fits provide better supervision for the network. We demonstrate the effectiveness of our approach in different settings, where 3D ground truth is scarce, or not available, and we consistently outperform the state-of-the-art model-based pose estimation approaches by significant margins. The project website with videos, results, and code can be found at https://seas.upenn.edu/~nkolot/projects/spin.

Large datasets are the cornerstone of recent advances in computer vision using deep learning. In contrast, existing human motion capture (mocap) datasets are small and the motions limited, hampering progress on learning models of human motion. While there are many different datasets available, they each use a different parameterization of the body, making it difficult to integrate them into a single meta dataset. To address this, we introduce AMASS, a large and varied database of human motion that unifies 15 different optical marker-based mocap datasets by representing them within a common framework and parameterization. We achieve this using a new method, MoSh++, that converts mocap data into realistic 3D human meshes represented by a rigged body model. Here we use SMPL [Loper et al., 2015], which is widely used and provides a standard skeletal representation as well as a fully rigged surface mesh. The method works for arbitrary marker sets, while recovering soft-tissue dynamics and realistic hand motion. We evaluate MoSh++ and tune its hyperparameters using a new dataset of 4D body scans that are jointly recorded with marker-based mocap. The consistent representation of AMASS makes it readily useful for animation, visualization, and generating training data for deep learning. Our dataset is significantly richer than previous human motion collections, having more than 40 hours of motion data, spanning over 300 subjects, more than 11000 motions, and will be publicly available to the research community.

Deep neural nets achieve state-of-the-art performance on the problem of optical flow estimation. Since optical flow is used in several safety-critical applications like self-driving cars, it is important to gain insights into the robustness of those techniques. Recently, it has been shown that adversarial attacks easily fool deep neural networks to misclassify objects. The robustness of optical flow networks to adversarial attacks, however, has not been studied so far. In this paper, we extend adversarial patch attacks to optical flow networks and show that such attacks can compromise their performance. We show that corrupting a small patch of less than 1% of the image size can significantly affect optical flow estimates. Our attacks lead to noisy flow estimates that extend significantly beyond the region of the attack, in many cases even completely erasing the motion of objects in the scene. While networks using an encoder-decoder architecture are very sensitive to these attacks, we found that networks using a spatial pyramid architecture are less affected. We analyse the success and failure of attacking both architectures by visualizing their feature maps and comparing them to classical optical flow techniques which are robust to these attacks. We also demonstrate that such attacks are practical by placing a printed pattern into real scenes.

Capturing human motion in natural scenarios means moving motion capture out of the lab and into the wild. Typical approaches rely on fixed, calibrated, cameras and reflective markers on the body, significantly limiting the motions that can be captured. To make motion capture truly unconstrained, we describe the first fully autonomous outdoor capture system based on flying vehicles. We use multiple micro-aerial-vehicles(MAVs), each equipped with a monocular RGB camera, an IMU, and a GPS receiver module. These detect the person, optimize their position, and localize themselves approximately. We then develop a markerless motion capture method that is suitable for this challenging scenario with a distant subject, viewed from above, with approximately calibrated and moving cameras. We combine multiple state-of-the-art 2D joint detectors with a 3D human body model and a powerful prior on human pose. We jointly optimize for 3D body pose and camera pose to robustly fit the 2D measurements. To our knowledge, this is the first successful demonstration of outdoor, full-body, markerless motion capture from autonomous flying vehicles.

We present the first method to perform automatic 3D pose, shape and texture capture of animals from images acquired in-the-wild. In particular, we focus on the problem of capturing 3D information about Grevy's zebras from a collection of images. The Grevy's zebra is one of the most endangered species in Africa, with only a few thousand individuals left. Capturing the shape and pose of these animals can provide biologists and conservationists with information about animal health and behavior. In contrast to research on human pose, shape and texture estimation, training data for endangered species is limited, the animals are in complex natural scenes with occlusion, they are naturally camouflaged, travel in herds, and look similar to each other. To overcome these challenges, we integrate the recent SMAL animal model into a network-based regression pipeline, which we train end-to-end on synthetically generated images with pose, shape, and background variation. Going beyond state-of-the-art methods for human shape and pose estimation, our method learns a shape space for zebras during training. Learning such a shape space from images using only a photometric loss is novel, and the approach can be used to learn shape in other settings with limited 3D supervision. Moreover, we couple 3D pose and shape prediction with the task of texture synthesis, obtaining a full texture map of the animal from a single image. We show that the predicted texture map allows a novel per-instance unsupervised optimization over the network features. This method, SMALST (SMAL with learned Shape and Texture) goes beyond previous work, which assumed manual keypoints and/or segmentation, to regress directly from pixels to 3D animal shape, pose and texture.

The individual shape of the human body, including the geometry of its articulated structure and the distribution of weight over that structure, influences the kinematics of a person’s movements. How sensitive is the visual system to inconsistencies between shape and motion introduced by retargeting motion from one person onto the shape of another? We used optical motion capture to record five pairs of male performers with large differences in body weight, while they pushed, lifted, and threw objects. From these data, we estimated both the kinematics of the actions as well as the performer’s individual body shape. To obtain consistent and inconsistent stimuli, we created animated avatars by combining the shape and motion estimates from either a single performer or from different performers. Using these stimuli we conducted three experiments in an immersive virtual reality environment. First, a group of participants detected which of two stimuli was inconsistent. Performance was very low, and results were only marginally significant. Next, a second group of participants rated perceived attractiveness, eeriness, and humanness of consistent and inconsistent stimuli, but these judgements of animation characteristics were not affected by consistency of the stimuli. Finally, a third group of participants rated properties of the objects rather than of the performers. Here, we found strong influences of shape-motion inconsistency on perceived weight and thrown distance of objects. This suggests that the visual system relies on its knowledge of shape and motion and that these components are assimilated into an altered perception of the action outcome. We propose that the visual system attempts to resist inconsistent interpretations of human animations. Actions involving object manipulations present an opportunity for the visual system to reinterpret the introduced inconsistencies as a change in the dynamics of an object rather than as an unexpected combination of body shape and body motion.

Creating metrically accurate avatars is important for many applications such as virtual clothing try-on, ergonomics, medicine, immersive social media, telepresence, and gaming. Creating avatars that precisely represent a particular individual is challenging however, due to the need for expensive 3D scanners, privacy issues with photographs or videos, and difficulty in making accurate tailoring measurements. We overcome these challenges by creating "The Virtual Caliper", which uses VR game controllers to make simple measurements. First, we establish what body measurements users can reliably make on their own body. We find several distance measurements to be good candidates and then verify that these are linearly related to 3D body shape as represented by the SMPL body model. The Virtual Caliper enables novice users to accurately measure themselves and create an avatar with their own body shape. We evaluate the metric accuracy relative to ground truth 3D body scan data, compare the method quantitatively to other avatar creation tools, and perform extensive perceptual studies. We also provide a software application to the community that enables novices to rapidly create avatars in fewer than five minutes. Not only is our approach more rapid than existing methods, it exports a metrically accurate 3D avatar model that is rigged and skinned.

We describe Human Mesh Recovery (HMR), an end-to-end framework for reconstructing a full 3D mesh of a human body from a single RGB image. In contrast to most current methods that compute 2D or 3D joint locations, we produce a richer and more useful mesh representation that is parameterized by shape and 3D joint angles. The main objective is to minimize the reprojection loss of keypoints, which allows our model to be trained using in-the-wild images that only have ground truth 2D annotations. However, the reprojection loss alone is highly underconstrained. In this work we address this problem by introducing an adversary trained to tell whether human body shape and pose parameters are real or not using a large database of 3D human meshes. We show that HMR can be trained with and without using any paired 2D-to-3D supervision. We do not rely on intermediate 2D keypoint detections and infer 3D pose and shape parameters directly from image pixels. Our model runs in real-time given a bounding box containing the person. We demonstrate our approach on various images in-the-wild and out-perform previous optimization-based methods that output 3D meshes and show competitive results on tasks such as 3D joint location estimation and part segmentation.

Animals are widespread in nature and the analysis of their shape and motion is important in many fields and industries. Modeling 3D animal shape, however, is difficult because the 3D scanning methods used to capture human shape are not applicable to wild animals or natural settings. Consequently, we propose a method to capture the detailed 3D shape of animals from images alone. The articulated and deformable nature of animals makes this problem extremely challenging, particularly in unconstrained environments with moving and uncalibrated cameras. To make this possible, we use a strong prior model of articulated animal shape that we fit to the image data. We then deform the animal shape in a canonical reference pose such that it matches image evidence when articulated and projected into multiple images. Our method extracts significantly more 3D shape detail than previous methods and is able to model new species, including the shape of an extinct animal, using only a few video frames. Additionally, the projected 3D shapes are accurate enough to facilitate the extraction of a realistic texture map from multiple frames.

Abstract Motion capture is often retargeted to new, and sometimes drastically different, characters. When the characters take on realistic human shapes, however, we become more sensitive to the motion looking right. This means adapting it to be consistent with the physical constraints imposed by different body shapes. We show how to take realistic 3D human shapes, approximate them using a simplified representation, and animate them so that they move realistically using physically‐based retargeting. We develop a novel spacetime optimization approach that learns and robustly adapts physical controllers to new bodies and constraints. The approach automatically adapts the motion of the mocap subject to the body shape of a target subject. This motion respects the physical properties of the new body and every body shape results in a different and appropriate movement. This makes it easy to create a varied set of motions from a single mocap sequence by simply varying the characters. In an interactive environment, successful retargeting requires adapting the motion to unexpected external forces. We achieve robustness to such forces using a novel LQR‐tree formulation. We show that the simulated motions look appropriate to each character's anatomy and their actions are robust to perturbations.

We demonstrate a novel deep neural network capable of reconstructing human full body pose in real-time from 6 Inertial Measurement Units (IMUs) worn on the user's body. In doing so, we address several difficult challenges. First, the problem is severely under-constrained as multiple pose parameters produce the same IMU orientations. Second, capturing IMU data in conjunction with ground-truth poses is expensive and difficult to do in many target application scenarios (e.g., outdoors). Third, modeling temporal dependencies through non-linear optimization has proven effective in prior work but makes real-time prediction infeasible. To address this important limitation, we learn the temporal pose priors using deep learning. To learn from sufficient data, we synthesize IMU data from motion capture datasets. A bi-directional RNN architecture leverages past and future information that is available at training time. At test time, we deploy the network in a sliding window fashion, retaining real time capabilities. To evaluate our method, we recorded DIP-IMU, a dataset consisting of 10 subjects wearing 17 IMUs for validation in 64 sequences with 330 000 time instants; this constitutes the largest IMU dataset publicly available. We quantitatively evaluate our approach on multiple datasets and show results from a real-time implementation. DIP-IMU and the code are available for research purposes. 1

The individual shape of the human body, including the geometry of its articulated structure and the distribution of weight over that structure, influences the kinematics of a person's movements. How sensitive is the visual system to inconsistencies between shape and motion introduced by retargeting motion from one person onto the shape of another? We used optical motion capture to record five pairs of male performers with large differences in body weight, while they pushed, lifted, and threw objects. Based on a set of 67 markers, we estimated both the kinematics of the actions as well as the performer's individual body shape. To obtain consistent and inconsistent stimuli, we created animated avatars by combining the shape and motion estimates from either a single performer or from different performers. In a virtual reality environment, observers rated the perceived weight or thrown distance of the objects. They were also asked to explicitly discriminate between consistent and hybrid stimuli. Observers were unable to accomplish the latter, but hybridization of shape and motion influenced their judgements of action outcome in systematic ways. Inconsistencies between shape and motion were assimilated into an altered perception of the action outcome.

While the ready availability of 3D scan data has influenced research throughout computer vision, less attention has focused on 4D data, that is 3D scans of moving non-rigid objects, captured over time. To be useful for vision research, such 4D scans need to be registered, or aligned, to a common topology. Consequently, extending mesh registration methods to 4D is important. Unfortunately, no ground-truth datasets are available for quantitative evaluation and comparison of 4D registration methods. To address this we create a novel dataset of high-resolution 4D scans of human subjects in motion, captured at 60 fps. We propose a new mesh registration method that uses both 3D geometry and texture information to register all scans in a sequence to a common reference topology. The approach exploits consistency in texture over both short and long time intervals and deals with temporal offsets between shape and texture capture. We show how using geometry alone results in significant errors in alignment when the motions are fast and non-rigid. We evaluate the accuracy of our registration and provide a dataset of 40,000 raw and aligned meshes. Dynamic FAUST extends the popular FAUST dataset to dynamic 4D data, and is available for research purposes at http://dfaust.is.tue.mpg.de.

Generative models of 3D human motion are often restricted to a small number of activities and can therefore not generalize well to novel movements or applications. In this work we propose a deep learning framework for human motion capture data that learns a generic representation from a large corpus of motion capture data and generalizes well to new, unseen, motions. Using an encoding-decoding network that learns to predict future 3D poses from the most recent past, we extract a feature representation of human motion. Most work on deep learning for sequence prediction focuses on video and speech. Since skeletal data has a different structure, we present and evaluate different network architectures that make different assumptions about time dependencies and limb correlations. To quantify the learned features, we use the output of different layers for action classification and visualize the receptive fields of the network units. Our method outperforms the recent state of the art in skeletal motion prediction even though these use action specific training data. Our results show that deep feedforward networks, trained from a generic mocap database, can successfully be used for feature extraction from human motion data and that this representation can be used as a foundation for classification and prediction.

Existing optical flow datasets are limited in size and variability due to the difficulty of capturing dense ground truth. In this paper, we tackle this problem by tracking pixels through densely sampled space-time volumes recorded with a high-speed video camera. Our model exploits the linearity of small motions and reasons about occlusions from multiple frames. Using our technique, we are able to establish accurate reference flow fields outside the laboratory in natural environments. Besides, we show how our predictions can be used to augment the input images with realistic motion blur. We demonstrate the quality of the produced flow fields on synthetic and real-world datasets. Finally, we collect a novel challenging optical flow dataset by applying our technique on data from a high-speed camera and analyze the performance of the state-of-the-art in optical flow under various levels of motion blur.

3D models provide a common ground for different representations of human bodies. In turn, robust 2D estimation has proven to be a powerful tool to obtain 3D fits in-the-wild. However, depending on the level of detail, it can be hard to impossible to acquire labeled data for training 2D estimators on large scale. We propose a hybrid approach to this problem: with an extended version of the recently introduced SMPLify method, we obtain high quality 3D body model fits for multiple human pose datasets. Human annotators solely sort good and bad fits. This procedure leads to an initial dataset, UP-3D, with rich annotations. With a comprehensive set of experiments, we show how this data can be used to train discriminative models that produce results with an unprecedented level of detail: our models predict 31 segments and 91 landmark locations on the body. Using the 91 landmark pose estimator, we present state-of-the art results for 3D human pose and shape estimation using an order of magnitude less training data and without assumptions about gender or pose in the fitting procedure. We show that UP-3D can be enhanced with these improved fits to grow in quantity and quality, which makes the system deployable on large scale. The data, code and models are available for research purposes.

Human motion modelling is a classical problem at the intersection of graphics and computer vision, with applications spanning human-computer interaction, motion synthesis, and motion prediction for virtual and augmented reality. Following the success of deep learning methods in several computer vision tasks, recent work has focused on using deep recurrent neural networks (RNNs) to model human motion, with the goal of learning time-dependent representations that perform tasks such as short-term motion prediction and long-term human motion synthesis. We examine recent work, with a focus on the evaluation methodologies commonly used in the literature, and show that, surprisingly, state of the art performance can be achieved by a simple baseline that does not attempt to model motion at all. We investigate this result, and analyze recent RNN methods by looking at the architectures, loss functions, and training procedures used in state-of-the-art approaches. We propose three changes to the standard RNN models typically used for human motion, which results in a simple and scalable RNN architecture that obtains state-of-the-art performance on human motion prediction.

We learn to compute opticalflow by combining a classical spatial-pyramid formulation with deep learning. This estimates large motions in a coarse-to-fine approach by warping one image of a pair at each pyramid level by the current flow estimate and computing an update to the flow. Instead of the standard minimization of an objective function at each pyramid level, we train one deep network per level to compute the flow update. Unlike the recent FlowNet approach, the networks do not need to deal with large motions; these are dealt with by the pyramid. This has several advantages. First, our Spatial Pyramid Network (SPyNet) is much simpler and 96% smaller than FlowNet in terms of model parameters. This makes it more efficient and appropriate for embedded applications. Second, since the flow at each pyramid level is small (<; 1 pixel), a convolutional approach applied to pairs of warped images is appropriate. Third, unlike FlowNet, the learned convolution filters appear similar to classical spatio-temporal filters, giving insight into the method and how to improve it. Our results are more accurate than FlowNet on most standard benchmarks, suggesting a new direction of combining classical flow methods with deep learning.

Dense 3D reconstruction from RGB images is a highly ill-posed problem due to occlusions, textureless or reflective surfaces, as well as other challenges. We propose object-level shape priors to address these ambiguities. Towards this goal, we formulate a probabilistic model that integrates multi-view image evidence with 3D shape information from multiple objects. Inference in this model yields a dense 3D reconstruction of the scene as well as the existence and precise 3D pose of the objects in it. Our approach is able to recover fine details not captured in the input shapes while defaulting to the input models in occluded regions where image evidence is weak. Due to its probabilistic nature, the approach is able to cope with the approximate geometry of the 3D models as well as input shapes that are not present in the scene. We evaluate the approach quantitatively on several challenging indoor and outdoor datasets.

Estimating human pose, shape, and motion from images and videos are fundamental challenges with many applications. Recent advances in 2D human pose estimation use large amounts of manually-labeled training data for learning convolutional neural networks (CNNs). Such data is time consuming to acquire and difficult to extend. Moreover, manual labeling of 3D pose, depth and motion is impractical. In this work we present SURREAL (Synthetic hUmans foR REAL tasks): a new large-scale dataset with synthetically-generated but realistic images of people rendered from 3D sequences of human motion capture data. We generate more than 6 million frames together with ground truth pose, depth maps, and segmentation masks. We show that CNNs trained on our synthetic dataset allow for accurate human depth estimation and human part segmentation in real RGB images. Our results and the new dataset open up new possibilities for advancing person analysis using cheap and large-scale synthetic data.

The optical flow of natural scenes is a combination of the motion of the observer and the independent motion of objects. Existing algorithms typically focus on either recovering motion and structure under the assumption of a purely static world or optical flow for general unconstrained scenes. We combine these approaches in an optical flow algorithm that estimates an explicit segmentation of moving objects from appearance and physical constraints. In static regions we take advantage of strong constraints to jointly estimate the camera motion and the 3D structure of the scene over multiple frames. This allows us to also regularize the structure instead of the motion. Our formulation uses a Plane+Parallax framework, which works even under small baselines, and reduces the motion estimation to a one-dimensional search problem, resulting in more accurate estimation. In moving regions the flow is treated as unconstrained, and computed with an existing optical flow method. The resulting Mostly-Rigid Flow (MR-Flow) method achieves state-of-the-art results on both the MPI-Sintel and KITTI-2015 benchmarks.

We address the problem of estimating human pose and body shape from 3D scans over time. Reliable estimation of 3D body shape is necessary for many applications including virtual try-on, health monitoring, and avatar creation for virtual reality. Scanning bodies in minimal clothing, however, presents a practical barrier to these applications. We address this problem by estimating body shape under clothing from a sequence of 3D scans. Previous methods that have exploited body models produce smooth shapes lacking personalized details. We contribute a new approach to recover a personalized shape of the person. The estimated shape deviates from a parametric model to fit the 3D scans. We demonstrate the method using high quality 4D data as well as sequences of visual hulls extracted from multi-view images. We also make available BUFF, a new 4D dataset that enables quantitative evaluation (http://buff.is.tue.mpg.de). Our method outperforms the state of the art in both pose estimation and shape estimation, qualitatively and quantitatively.

There has been significant work on learning realistic, articulated, 3D models of the human body. In contrast, there are few such models of animals, despite many applications. The main challenge is that animals are much less cooperative than humans. The best human body models are learned from thousands of 3D scans of people in specific poses, which is infeasible with live animals. Consequently, we learn our model from a small set of 3D scans of toy figurines in arbitrary poses. We employ a novel part-based shape model to compute an initial registration to the scans. We then normalize their pose, learn a statistical shape model, and refine the registrations and the model together. In this way, we accurately align animal scans from different quadruped families with very different shapes and poses. With the registration to a common template we learn a shape space representing animals including lions, cats, dogs, horses, cows and hippos. Animal shapes can be sampled from the model, posed, animated, and fit to data. We demonstrate generalization by fitting it to images of real animals including species not seen in training.

We address the problem of making human motion capture in the wild more practical by using a small set of inertial sensors attached to the body. Since the problem is heavily under-constrained, previous methods either use a large number of sensors, which is intrusive, or they require additional video input. We take a different approach and constrain the problem by: (i) making use of a realistic statistical body model that includes anthropometric constraints and (ii) using a joint optimization framework to fit the model to orientation and acceleration measurements over multiple frames. The resulting tracker Sparse Inertial Poser (SIP) enables motion capture using only 6 sensors (attached to the wrists, lower legs, back and head) and works for arbitrary human motions. Experiments on the recently released TNT15 dataset show that, using the same number of sensors, SIP achieves higher accuracy than the dataset baseline without using any video data. We further demonstrate the effectiveness of SIP on newly recorded challenging motions in outdoor scenarios such as climbing or jumping over a wall.

Humans move their hands and bodies together to communicate and solve tasks. Capturing and replicating such coordinated activity is critical for virtual characters that behave realistically. Surprisingly, most methods treat the 3D modeling and tracking of bodies and hands separately. Here we formulate a model of hands and bodies interacting together and fit it to full-body 4D sequences. When scanning or capturing the full body in 3D, hands are small and often partially occluded, making their shape and pose hard to recover. To cope with low-resolution, occlusion, and noise, we develop a new model called MANO ( hand Model with Articulated and Non-rigid defOrmations ). MANO is learned from around 1000 high-resolution 3D scans of hands of 31 subjects in a wide variety of hand poses. The model is realistic, low-dimensional, captures non-rigid shape changes with pose, is compatible with standard graphics packages, and can fit any human hand. MANO provides a compact mapping from hand poses to pose blend shape corrections and a linear manifold of pose synergies. We attach MANO to a standard parameterized 3D body shape model (SMPL), resulting in a fully articulated body and hand model (SMPL+H). We illustrate SMPL+H by fitting complex, natural, activities of subjects captured with a 4D scanner. The fitting is fully automatic and results in full body models that move naturally with detailed hand motions and a realism not seen before in full body performance capture. The models and data are freely available for research purposes at http://mano.is.tue.mpg.de.

Data driven models of human poses and soft-tissue deformations can produce very realistic results, but they only model the visible surface of the human body and cannot create skin deformation due to interactions with the environment. Physical simulations can generalize to external forces, but their parameters are difficult to control. In this paper, we present a layered volumetric human body model learned from data. Our model is composed of a data-driven inner layer and a physics-based external layer. The inner layer is driven with a volumetric statistical body model (VSMPL). The soft tissue layer consists of a tetrahedral mesh that is driven using the finite element method (FEM). Model parameters, namely the segmentation of the body into layers and the soft tissue elasticity, are learned directly from 4D registrations of humans exhibiting soft tissue deformations. The learned two layer model is a realistic full-body avatar that generalizes to novel motions and external forces. Experiments show that the resulting avatars produce realistic results on held out sequences and react to external forces. Moreover, the model supports the retargeting of physical properties from one avatar when they share the same topology.

The field of 3D face modeling has a large gap between high-end and low-end methods. At the high end, the best facial animation is indistinguishable from real humans, but this comes at the cost of extensive manual labor. At the low end, face capture from consumer depth sensors relies on 3D face models that are not expressive enough to capture the variability in natural facial shape and expression. We seek a middle ground by learning a facial model from thousands of accurately aligned 3D scans. Our FLAME model (Faces Learned with an Articulated Model and Expressions) is designed to work with existing graphics software and be easy to fit to data. FLAME uses a linear shape space trained from 3800 scans of human heads. FLAME combines this linear shape space with an articulated jaw, neck, and eyeballs, pose-dependent corrective blendshapes, and additional global expression blendshapes. The pose and expression dependent articulations are learned from 4D face sequences in the D3DFACS dataset along with additional 4D sequences. We accurately register a template mesh to the scan sequences and make the D3DFACS registrations available for research purposes. In total the model is trained from over 33, 000 scans. FLAME is low-dimensional but more expressive than the FaceWarehouse model and the Basel Face Model. We compare FLAME to these models by fitting them to static 3D scans and 4D sequences using the same optimization method. FLAME is significantly more accurate and is available for research purposes (http://flame.is.tue.mpg.de).

Designing and simulating realistic clothing is challenging. Previous methods addressing the capture of clothing from 3D scans have been limited to single garments and simple motions, lack detail, or require specialized texture patterns. Here we address the problem of capturing regular clothing on fully dressed people in motion. People typically wear multiple pieces of clothing at a time. To estimate the shape of such clothing, track it over time, and render it believably, each garment must be segmented from the others and the body. Our ClothCap approach uses a new multi-part 3D model of clothed bodies, automatically segments each piece of clothing, estimates the minimally clothed body shape and pose under the clothing, and tracks the 3D deformations of the clothing over time. We estimate the garments and their motion from 4D scans; that is, high-resolution 3D scans of the subject in motion at 60 fps. ClothCap is able to capture a clothed person in motion, extract their clothing, and retarget the clothing to new body shapes; this provides a step towards virtual try-on.

Occlusion boundaries contain rich perceptual information about the underlying scene structure. They also provide important cues in many visual perception tasks such as scene understanding, object recognition, and segmentation. In this paper, we improve occlusion boundary detection via enhanced exploration of contextual information (e.g., local structural boundary patterns, observations from surrounding regions, and temporal context), and in doing so develop a novel approach based on convolutional neural networks (CNNs) and conditional random fields (CRFs). Experimental results demonstrate that our detector significantly outperforms the state-of-the-art (e.g., improving the F-measure from 0.62 to 0.71 on the commonly used CMU benchmark). Last but not least, we empirically assess the roles of several important components of the proposed detector, so as to validate the rationale behind this approach.

Existing optical flow methods make generic, spatially homogeneous, assumptions about the spatial structure of the flow. In reality, optical flow varies across an image depending on object class. Simply put, different objects move differently. Here we exploit recent advances in static semantic scene segmentation to segment the image into objects of different types. We define different models of image motion in these regions depending on the type of object. For example, we model the motion on roads with homographies, vegetation with spatially smooth flow, and independently moving objects like cars and planes with affine motion plus deviations. We then pose the flow estimation problem using a novel formulation of localized layers, which addresses limitations of traditional layered models for dealing with complex scene motion. Our semantic flow method achieves the lowest error of any published monocular method in the KITTI-2015 flow benchmark and produces qualitatively better flow and segmentation than recent top methods on a wide range of natural videos.

Video object segmentation is challenging due to fast moving objects, deforming shapes, and cluttered backgrounds. Optical flow can be used to propagate an object segmentation over time but, unfortunately, flow is often inaccurate, particularly around object boundaries. Such boundaries are precisely where we want our segmentation to be accurate. To obtain accurate segmentation across time, we propose an efficient algorithm that considers video segmentation and optical flow estimation simultaneously. For video segmentation, we formulate a principled, multiscale, spatio-temporal objective function that uses optical flow to propagate information between frames. For optical flow estimation, particularly at object boundaries, we compute the flow independently in the segmented regions and recompose the results. We call the process object flow and demonstrate the effectiveness of jointly optimizing optical flow and video segmentation using an iterative scheme. Experiments on the SegTrack v2 and Youtube-Objects datasets show that the proposed algorithm performs favorably against the other state-of-the-art methods.

In this paper, we propose a non-local structured prior for volumetric multi-view 3D reconstruction. Towards this goal, we present a novel Markov random field model based on ray potentials in which assumptions about large 3D surface patches such as planarity or Manhattan world constraints can be efficiently encoded as probabilistic priors. We further derive an inference algorithm that reasons jointly about voxels, pixels and image segments, and estimates marginal distributions of appearance, occupancy, depth, normals and planarity. Key to tractable inference is a novel hybrid representation that spans both voxel and pixel space and that integrates non-local information from 2D image segmentations in a principled way. We compare our non-local prior to commonly employed local smoothness assumptions and a variety of state-of-the-art volumetric reconstruction baselines on challenging outdoor scenes with textureless and reflective surfaces. Our experiments indicate that regularizing over larger distances has the potential to resolve ambiguities where local regularizers fail.

In large datasets, manual data verification is impossible, and we must expect the number of outliers to increase with data size. While principal component analysis (PCA) can reduce data size, and scalable solutions exist, it is well-known that outliers can arbitrarily corrupt the results. Unfortunately, state-of-the-art approaches for robust PCA are not scalable. We note that in a zero-mean dataset, each observation spans a one-dimensional subspace, giving a point on the Grassmann manifold. We show that the average subspace corresponds to the leading principal component for Gaussian data. We provide a simple algorithm for computing this Grassmann Average ( GA), and show that the subspace estimate is less sensitive to outliers than PCA for general distributions. Because averages can be efficiently computed, we immediately gain scalability. We exploit robust averaging to formulate the Robust Grassmann Average (RGA) as a form of robust PCA. The resulting Trimmed Grassmann Average ( TGA) is appropriate for computer vision because it is robust to pixel outliers. The algorithm has linear computational complexity and minimal memory requirements. We demonstrate TGA for background modeling, video restoration, and shadow removal. We show scalability by performing robust PCA on the entire Star Wars IV movie; a task beyond any current method. Source code is available online.

Realistic, metrically accurate, 3D human avatars are useful for games, shopping, virtual reality, and health applications. Such avatars are not in wide use because solutions for creating them from high-end scanners, low-cost range cameras, and tailoring measurements all have limitations. Here we propose a simple solution and show that it is surprisingly accurate. We use crowdsourcing to generate attribute ratings of 3D body shapes corresponding to standard linguistic descriptions of 3D shape. We then learn a linear function relating these ratings to 3D human shape parameters. Given an image of a new body, we again turn to the crowd for ratings of the body shape. The collection of linguistic ratings of a photograph provides remarkably strong constraints on the metric 3D shape. We call the process crowdshaping and show that our Body Talk system produces shapes that are perceptually indistinguishable from bodies created from high-resolution scans and that the metric accuracy is sufficient for many tasks. This makes body "scanning" practical without a scanner, opening up new applications including database search, visualization, and extracting avatars from books.

This paper presents a novel probabilistic foundation for volumetric 3D reconstruction. We formulate the problem as inference in a Markov random field, which accurately captures the dependencies between the occupancy and appearance of each voxel, given all input images. Our main contribution is an approximate highly parallelized discrete-continuous inference algorithm to compute the marginal distributions of each voxel's occupancy and appearance. In contrast to the MAP solution, marginals encode the underlying uncertainty and ambiguity in the reconstruction. Moreover, the proposed algorithm allows for a Bayes optimal prediction with respect to a natural reconstruction loss. We compare our method to two state-of-the-art volumetric reconstruction algorithms on three challenging aerial datasets with LIDAR ground truth. Our experiments demonstrate that the proposed algorithm compares favorably in terms of reconstruction accuracy and the ability to expose reconstruction uncertainty.

We investigated the influence of body shape and pose on the perception of physical strength and social power for male virtual characters. In the first experiment, participants judged the physical strength of varying body shapes, derived from a statistical 3D body model. Based on these ratings, we determined three body shapes (weak, average, and strong) and animated them with a set of power poses for the second experiment. Participants rated how strong or powerful they perceived virtual characters of varying body shapes that were displayed in different poses. Our results show that perception of physical strength was mainly driven by the shape of the body. However, the social attribute of power was influenced by an interaction between pose and shape. Specifically, the effect of pose on power ratings was greater for weak body shapes. These results demonstrate that a character with a weak shape can be perceived as more powerful when in a high-power pose.

Estimating 3D human pose from 2D joint locations is central to the analysis of people in images and video. To address the fact that the problem is inherently ill posed, many methods impose a prior over human poses. Unfortunately these priors admit invalid poses because they do not model how joint-limits vary with pose. Here we make two key contributions. First, we collect a motion capture dataset that explores a wide range of human poses. From this we learn a pose-dependent model of joint limits that forms our prior. Both dataset and prior are available for research purposes. Second, we define a general parametrization of body pose and a new, multi-stage, method to estimate 3D pose from 2D joint locations using an over-complete dictionary of poses. Our method shows good generalization while avoiding impossible poses. We quantitatively compare our method with recent work and show state-of-the-art results on 2D to 3D pose estimation using the CMU mocap dataset. We also show superior results using manual annotations on real images and automatic detections on the Leeds sports pose dataset.

We address the elusive goal of estimating optical flow both accurately and efficiently by adopting a sparse-to-dense approach. Given a set of sparse matches, we regress to dense optical flow using a learned set of full-frame basis flow fields. We learn the principal components of natural flow fields using flow computed from four Hollywood movies. Optical flow fields are then compactly approximated as a weighted sum of the basis flow fields. Our new PCA-Flow algorithm robustly estimates these weights from sparse feature matches. The method runs in under 200ms/frame on the MPI-Sintel dataset using a single CPU and is more accurate and significantly faster than popular methods such as LDOF and Classic+NL. For some applications, however, the results are too smooth. Consequently, we develop a novel sparse layered flow method in which each layer is represented by PCA-Flow. Unlike existing layered methods, estimation is fast because it uses only sparse matches. We combine information from different layers into a dense flow field using an image-aware MRF. The resulting PCA-Layers method runs in 3.2s/frame, is significantly more accurate than PCA-Flow, and achieves state-of-the-art performance in occluded regions on MPI-Sintel.

We propose a new 3D model of the human body that is both realistic and part-based. The body is represented by a graphical model in which nodes of the graph correspond to body parts that can independently translate and rotate in 3D and deform to represent different body shapes and to capture pose-dependent shape variations. Pairwise potentials define a “stitching cost” for pulling the limbs apart, giving rise to the stitched puppet (SP) model. Unlike existing realistic 3D body models, the distributed representation facilitates inference by allowing the model to more effectively explore the space of poses, much like existing 2D pictorial structures models. We infer pose and body shape using a form of particle-based max-product belief propagation. This gives SP the realism of recent 3D body models with the computational advantages of part-based models. We apply SP to two challenging problems involving estimating human shape and pose from 3D data. The first is the FAUST mesh alignment challenge, where ours is the first method to successfully align all 3D meshes with no pose prior. The second involves estimating pose and shape from crude visual hull representations of complex body movements.

We accurately estimate the 3D geometry and appearance of the human body from a monocular RGB-D sequence of a user moving freely in front of the sensor. Range data in each frame is first brought into alignment with a multi-resolution 3D body model in a coarse-to-fine process. The method then uses geometry and image texture over time to obtain accurate shape, pose, and appearance information despite unconstrained motion, partial views, varying resolution, occlusion, and soft tissue deformation. Our novel body model has variable shape detail, allowing it to capture faces with a high-resolution deformable head model and body shape with lower-resolution. Finally we combine range data from an entire sequence to estimate a high-resolution displacement map that captures fine shape details. We compare our recovered models with high-resolution scans from a professional system and with avatars created by a commercial product. We extract accurate 3D avatars from challenging motion sequences and even capture soft tissue dynamics.

We formulate the estimation of dense depth maps from video sequences as a problem of intrinsic image estimation. Our approach synergistically integrates the estimation of multiple intrinsic images including depth, albedo, shading, optical flow, and surface contours. We build upon an example-based framework for depth estimation that uses label transfer from a database of RGB and depth pairs. We combine this with a method that extracts consistent albedo and shading from video. In contrast to raw RGB values, albedo and shading provide a richer, more physical, foundation for depth transfer. Additionally we train a new contour detector to predict surface boundaries from albedo, shading, and pixel values and use this to improve the estimation of depth boundaries. We also integrate sparse structure from motion with our method to improve the metric accuracy of the estimated depth maps. We evaluate our Intrinsic Depth method quantitatively by estimating depth from videos in the NYU RGB-D and SUN3D datasets. We find that combining the estimation of multiple intrinsic images improves depth estimation relative to the baseline method.

Increased emphasis on circuit level activity in the brain makes it necessary to have methods to visualize and evaluate large-scale ensemble activity beyond that revealed by raster-histograms or pairwise correlations. We present a method to evaluate the relative similarity of neural spiking patterns by combining spike train distance metrics with dimensionality reduction. Spike train distance metrics provide an estimate of similarity between activity patterns at multiple temporal resolutions. Vectors of pair-wise distances are used to represent the intrinsic relationships between multiple activity patterns at the level of single units or neuronal ensembles. Dimensionality reduction is then used to project the data into concise representations suitable for clustering analysis as well as exploratory visualization. Algorithm performance and robustness are evaluated using multielectrode ensemble activity data recorded in behaving primates. We demonstrate how spike train SIMilarity space (SSIMS) analysis captures the relationship between goal directions for an eight-directional reaching task and successfully segregates grasp types in a 3D grasping task in the absence of kinematic information. The algorithm enables exploration of virtually any type of neural spiking (time series) data, providing similarity-based clustering of neural activity states with minimal assumptions about potential information encoding models.

We present a learned model of human body shape and pose-dependent shape variation that is more accurate than previous models and is compatible with existing graphics pipelines. Our Skinned Multi-Person Linear model (SMPL) is a skinned vertex-based model that accurately represents a wide variety of body shapes in natural human poses. The parameters of the model are learned from data including the rest pose template, blend weights, pose-dependent blend shapes, identity-dependent blend shapes, and a regressor from vertices to joint locations. Unlike previous models, the pose-dependent blend shapes are a linear function of the elements of the pose rotation matrices. This simple formulation enables training the entire model from a relatively large number of aligned 3D meshes of different people in different poses. We quantitatively evaluate variants of SMPL using linear or dual-quaternion blend skinning and show that both are more accurate than a Blend-SCAPE model trained on the same data. We also extend SMPL to realistically model dynamic soft-tissue deformations. Because it is based on blend skinning, SMPL is compatible with existing rendering engines and we make it available for research purposes.

To look human, digital full-body avatars need to have soft-tissue deformations like those of real people. We learn a model of soft-tissue deformations from examples using a high-resolution 4D capture system and a method that accurately registers a template mesh to sequences of 3D scans. Using over 40,000 scans of ten subjects, we learn how soft-tissue motion causes mesh triangles to deform relative to a base 3D body model. Our Dyna model uses a low-dimensional linear subspace to approximate soft-tissue deformation and relates the subspace coefficients to the changing pose of the body. Dyna uses a second-order auto-regressive model that predicts soft-tissue deformations based on previous deformations, the velocity and acceleration of the body, and the angular velocities and accelerations of the limbs. Dyna also models how deformations vary with a person's body mass index (BMI), producing different deformations for people with different shapes. Dyna realistically represents the dynamics of soft tissue for previously unseen subjects and motions. We provide tools for animators to modify the deformations and apply them to new stylized characters.

New scanning technologies are increasing the importance of 3D mesh data and the need for algorithms that can reliably align it. Surface registration is important for building full 3D models from partial scans, creating statistical shape models, shape retrieval, and tracking. The problem is particularly challenging for non-rigid and articulated objects like human bodies. While the challenges of real-world data registration are not present in existing synthetic datasets, establishing ground-truth correspondences for real 3D scans is difficult. We address this with a novel mesh registration technique that combines 3D shape and appearance information to produce high-quality alignments. We define a new dataset called FAUST that contains 300 scans of 10 people in a wide range of poses together with an evaluation methodology. To achieve accurate registration, we paint the subjects with high-frequency textures and use an extensive validation process to ensure accurate ground truth. We find that current shape registration methods have trouble with this real-world data. The dataset and evaluation website are available for research purposes at http://faust.is.tue.mpg.de.

We consider the intersection of two research fields: transfer learning and statistics on manifolds. In particular, we consider, for manifold-valued data, transfer learning of tangent-space models such as Gaussians distributions, PCA, regression, or classifiers. Though one would hope to simply use ordinary Rn-transfer learning ideas, the manifold structure prevents it. We overcome this by basing our method on inner-product-preserving parallel transport, a well-known tool widely used in other problems of statistics on manifolds in computer vision. At first, this straight-forward idea seems to suffer from an obvious shortcoming: Transporting large datasets is prohibitively expensive, hindering scalability. Fortunately, with our approach, we never transport data. Rather, we show how the statistical models themselves can be transported, and prove that for the tangent-space models above, the transport "commutes" with learning. Consequently, our compact framework, ap- plicable to a large class of manifolds, is not restricted by the size of either the training or test sets. We demonstrate the approach by transferring PCA and logistic-regression models of real-world data involving 3D shapes and image descriptors.

As the collection of large datasets becomes increasingly automated, the occurrence of outliers will increase -- "big data" implies "big outliers". While principal component analysis (PCA) is often used to reduce the size of data, and scalable solutions exist, it is well-known that outliers can arbitrarily corrupt the results. Unfortunately, state-of-the-art approaches for robust PCA do not scale beyond small-to-medium sized datasets. To address this, we introduce the Grassmann Average (GA), which expresses dimensionality reduction as an average of the subspaces spanned by the data. Because averages can be efficiently computed, we immediately gain scalability. GA is inherently more robust than PCA, but we show that they coincide for Gaussian data. We exploit that averages can be made robust to formulate the Robust Grassmann Average (RGA) as a form of robust PCA. Robustness can be with respect to vectors (subspaces) or elements of vectors, we focus on the latter and use a trimmed average. The resulting Trimmed Grassmann Average (TGA) is particularly appropriate for computer vision because it is robust to pixel outliers. The algorithm has low computational complexity and minimal memory requirements, making it scalable to "big noisy data." We demonstrate TGA for background modeling, video restoration, and shadow removal. We show scalability by performing robust PCA on the entire Star Wars IV movie.

In applications of graphical models arising in domains such as computer vision and signal processing, we often seek the most likely configurations of high-dimensional, continuous variables. We develop a particle-based max-product algorithm which maintains a diverse set of posterior mode hypotheses, and is robust to initialization. At each iteration, the set of hypotheses at each node is augmented via stochastic proposals, and then reduced via an efficient selection algorithm. The integer program underlying our optimization-based particle selection minimizes errors in subsequent max-product message updates. This objective automatically encourages diversity in the maintained hypotheses, without requiring tuning of application-specific distances among hypotheses. By avoiding the stochastic resampling steps underlying particle sum-product algorithms, we also avoid common degeneracies where particles collapse onto a single hypothesis. Our approach significantly outperforms previous particle-based algorithms in experiments focusing on the estimation of human pose from single images.

Extracting anthropometric or tailoring measurements from 3D human body scans is important for applications such as virtual try-on, custom clothing, and online sizing. Existing commercial solutions identify anatomical landmarks on high-resolution 3D scans and then compute distances or circumferences on the scan. Landmark detection is sensitive to acquisition noise (e.g. holes) and these methods require subjects to adopt a specific pose. In contrast, we propose a solution we call model-based anthropometry. We fit a deformable 3D body model to scan data in one or more poses; this model-based fitting is robust to scan noise. This brings the scan into registration with a database of registered body scans. Then, we extract features from the registered model (rather than from the scan); these include, limb lengths, circumferences, and statistical features of global shape. Finally, we learn a mapping from these features to measurements using regularized linear regression. We perform an extensive evaluation using the CAESAR dataset and demonstrate that the accuracy of our method outperforms state-of-the-art methods.

The accuracy of optical flow estimation algorithms has been improving steadily as evidenced by results on the Middlebury optical flow benchmark. The typical formulation, however, has changed little since the work of Horn and Schunck. We attempt to uncover what has made recent advances possible through a thorough analysis of how the objective function, the optimization method, and modern implementation practices influence accuracy. We discover that “classical” flow formulations perform surprisingly well when combined with modern optimization and implementation techniques. One key implementation detail is the median filtering of intermediate flow fields during optimization. While this improves the robustness of classical methods it actually leads to higher energy solutions, meaning that these methods are not optimizing the original objective function. To understand the principles behind this phenomenon, we derive a new objective function that formalizes the median filtering heuristic. This objective function includes a non-local smoothness term that robustly integrates flow estimates over large spatial neighborhoods. By modifying this new term to include information about flow and image boundaries we develop a method that can better preserve motion details. To take advantage of the trend towards video in wide-screen format, we further introduce an asymmetric pyramid downsampling scheme that enables the estimation of longer range horizontal motions. The methods are evaluated on the Middlebury, MPI Sintel, and KITTI datasets using the same parameter settings.

The goal of this research was to investigate women's sensitivity to changes in their perceived weight by altering the body mass index (BMI) of the participants' personalized avatars displayed on a large-screen immersive display. We created the personalized avatars with a full-body 3D scanner that records the participants' body geometry and texture. We altered the weight of the personalized avatars to produce changes in BMI while keeping height, arm length, and inseam fixed and exploited the correlation between body geometry and anthropometric measurements encapsulated in a statistical body shape model created from thousands of body scans. In a 2 × 2 psychophysical experiment, we investigated the relative importance of visual cues, namely shape (own shape vs. an average female body shape with equivalent height and BMI to the participant) and texture (own photo-realistic texture or checkerboard pattern texture) on the ability to accurately perceive own current body weight (by asking the participant, “Is it the same weight as you?”). Our results indicate that shape (where height and BMI are fixed) had little effect on the perception of body weight. Interestingly, the participants perceived their body weight veridically when they saw their own photo-realistic texture. As compared to avatars with photo-realistic texture, the avatars with checkerboard texture needed to be significantly thinner in order to represent the participants' current weight. This suggests that in general the avatars with checkerboard texture appeared bigger. The range that the participants accepted as their own current weight was approximately a 0.83% to − 6.05% BMI change tolerance range around their perceived weight. Both the shape and the texture had an effect on the reported similarity of the body parts and the whole avatar to the participant's body. This work has implications for new measures for patients with body image disorders, as well as researchers interested in creating personalized avatars for games, training applications, or virtual reality.

Marker-based motion capture (mocap) is widely criticized as producing lifeless animations. We argue that important information about body surface motion is present in standard marker sets but is lost in extracting a skeleton. We demonstrate a new approach called MoSh (Motion and Shape capture), that automatically extracts this detail from mocap data. MoSh estimates body shape and pose together using sparse marker data by exploiting a parametric model of the human body. In contrast to previous work, MoSh solves for the marker locations relative to the body and estimates accurate body shape directly from the markers without the use of 3D scans; this effectively turns a mocap system into an approximate body scanner. MoSh is able to capture soft tissue motions directly from markers by allowing body shape to vary over time. We evaluate the effect of different marker sets on pose and shape accuracy and propose a new sparse marker set for capturing soft-tissue motion. We illustrate MoSh by recovering body shape, pose, and soft-tissue motion from archival mocap data and using this to produce animations with subtlety and realism. We also show soft-tissue motion retargeting to new characters and show how to magnify the 3D deformations of soft tissue to create animations with appealing exaggerations.

Modeling how the human body deforms during breathing is important for the realistic animation of lifelike 3D avatars. We learn a model of body shape deformations due to breathing for different breathing types and provide simple animation controls to render lifelike breathing regardless of body shape. We capture and align high-resolution 3D scans of 58 human subjects. We compute deviations from each subject's mean shape during breathing, and study the statistics of such shape changes for different genders, body shapes, and breathing types. We use the volume of the registered scans as a proxy for lung volume and learn a novel non-linear model relating volume and breathing type to 3D shape deformations and pose changes. We then augment a SCAPE body model so that body shape is determined by identity, pose, and the parameters of the breathing model. These parameters provide an intuitive interface with which animators can synthesize 3D human avatars with realistic breathing motions. We also develop a novel interface for animating breathing using a spirometer , which measures the changes in breathing volume of a "breath actor".

Layered models allow scene segmentation and motion estimation to be formulated together and to inform one another. Traditional layered motion methods, however, employ fairly weak models of scene structure, relying on locally connected Ising/Potts models which have limited ability to capture long-range correlations in natural scenes. To address this, we formulate a fully-connected layered model that enables global reasoning about the complicated segmentations of real objects. Optimization with fully-connected graphical models is challenging, and our inference algorithm leverages recent work on efficient mean field updates for fully-connected conditional random fields. These methods can be implemented efficiently using high-dimensional Gaussian filtering. We combine these ideas with a layered flow model, and find that the long-range connections greatly improve segmentation into figure-ground layers when compared with locally connected MRF models. Experiments on several benchmark datasets show that the method can recover fine structures and large occlusion regions, with good flow accuracy and much lower computational cost than previous locally-connected layered models.

Although action recognition in videos is widely studied, current methods often fail on real-world datasets. Many recent approaches improve accuracy and robustness to cope with challenging video sequences, but it is often unclear what affects the results most. This paper attempts to provide insights based on a systematic performance evaluation using thoroughly-annotated data of human actions. We annotate human Joints for the HMDB dataset (J-HMDB). This annotation can be used to derive ground truth optical flow and segmentation. We evaluate current methods using this dataset and systematically replace the output of various algorithms with ground truth. This enables us to discover what is important - for example, should we work on improving flow algorithms, estimating human bounding boxes, or enabling pose estimation? In summary, we find that high-level pose features greatly outperform low/mid level features, in particular, pose over time is critical, but current pose estimation algorithms are not yet reliable enough to provide this information. We also find that the accuracy of a top-performing action recognition framework can be greatly increased by refining the underlying low/mid level features, this suggests it is important to improve optical flow and human detection algorithms. Our analysis and J-HMDB dataset should facilitate a deeper understanding of action recognition algorithms.

We address the problem of upper-body human pose estimation in uncontrolled monocular video sequences, without manual initialization. Most current methods focus on isolated video frames and often fail to correctly localize arms and hands. Inferring pose over a video sequence is advantageous because poses of people in adjacent frames exhibit properties of smooth variation due to the nature of human and camera motion. To exploit this, previous methods have used prior knowledge about distinctive actions or generic temporal priors combined with static image likelihoods to track people in motion. Here we take a different approach based on a simple observation: Information about how a person moves from frame to frame is present in the optical flow field. We develop an approach for tracking articulated motions that "links" articulated shape models of people in adjacent frames through the dense optical flow. Key to this approach is a 2D shape model of the body that we use to compute how the body moves over time. The resulting "flowing puppets" provide a way of integrating image evidence across frames to improve pose inference. We apply our method on a challenging dataset of TV video sequences and show state-of-the-art performance.

Layered models provide a compelling approach for estimating image motion and segmenting moving scenes. Previous methods, however, have failed to capture the structure of complex scenes, provide precise object boundaries, effectively estimate the number of layers in a scene, or robustly determine the depth order of the layers. Furthermore, previous methods have focused on optical flow between pairs of frames rather than longer sequences. We show that image sequences with more frames are needed to resolve ambiguities in depth ordering at occlusion boundaries; temporal layer constancy makes this feasible. Our generative model of image sequences is rich but difficult to optimize with traditional gradient descent methods. We propose a novel discrete approximation of the continuous objective in terms of a sequence of depth-ordered MRFs and extend graph-cut optimization methods with new “moves” that make joint layer segmentation and motion estimation feasible. Our optimizer, which mixes discrete and continuous optimization, automatically determines the number of layers and reasons about their depth ordering. We demonstrate the value of layered models, our optimization strategy, and the use of more than two frames on both the Middlebury optical flow benchmark and the MIT layer segmentation benchmark.

Pictorial Structures (PS) define a probabilistic model of 2D articulated objects in images. Typical PS models assume an object can be represented by a set of rigid parts connected with pairwise constraints that define the prior probability of part configurations. These models are widely used to represent non-rigid articulated objects such as humans and animals despite the fact that such objects have parts that deform non-rigidly. Here we define a new Deformable Structures (DS) model that is a natural extension of previous PS models and that captures the non-rigid shape deformation of the parts. Each part in a DS model is represented by a low-dimensional shape deformation space and pairwise potentials between parts capture how the shape varies with pose and the shape of neighboring parts. A key advantage of such a model is that it more accurately models object boundaries. This enables image likelihood models that are more discriminative than previous PS likelihoods. This likelihood is learned using training imagery annotated using a DS “puppet.” We focus on a human DS model learned from 2D projections of a realistic 3D human body model and use it to infer human poses in images using a form of non-parametric belief propagation.

We develop a method for discovering the parts of an articulated object from aligned meshes capturing various three-dimensional (3D) poses. We adapt the distance dependent Chinese restaurant process (ddCRP) to allow nonparametric discovery of a potentially unbounded number of parts, while simultaneously guaranteeing a spatially connected segmentation. To allow analysis of datasets in which object instances have varying shapes, we model part variability across poses via affine transformations. By placing a matrix normal-inverse-Wishart prior on these affine transformations, we develop a ddCRP Gibbs sampler which tractably marginalizes over transformation uncertainty. Analyzing a dataset of humans captured in dozens of poses, we infer parts which provide quantitatively better motion predictions than conventional clustering methods.

Multi-metric learning techniques learn local metric tensors in different parts of a feature space. With such an approach, even simple classifiers can be competitive with the state-of-the-art because the distance measure locally adapts to the structure of the data. The learned distance measure is, however, non-metric, which has prevented multi-metric learning from generalizing to tasks such as dimensionality reduction and regression in a principled way. We prove that, with appropriate changes, multi-metric learning corresponds to learning the structure of a Riemannian manifold. We then show that this structure gives us a principled way to perform dimensionality reduction and regression according to the learned metrics. Algorithmically, we provide the first practical algorithm for computing geodesics according to the learned metrics, as well as algorithms for computing exponential and logarithmic maps on the Riemannian manifold. Together, these tools let many Euclidean algorithms take advantage of multi-metric learning. We illustrate the approach on regression and dimensionality reduction tasks that involve predicting measurements of the human body from shape data.

We formulate the problem of 3D human pose estimation and tracking as one of inference in a graphical model. Unlike traditional kinematic tree representations, our model of the body is a collection of loosely-connected body-parts. In particular, we model the body using an undirected graphical model in which nodes correspond to parts and edges to kinematic, penetration, and temporal constraints imposed by the joints and the world. These constraints are encoded using pair-wise statistical distributions, that are learned from motion-capture training data. Human pose and motion estimation is formulated as inference in this graphical model and is solved using Particle Message Passing ( PaMPas ). PaMPas is a form of non-parametric belief propagation that uses a variation of particle filtering that can be applied over a general graphical model with loops. The loose-limbed model and decentralized graph structure allow us to incorporate information from “bottom-up” visual cues, such as limb and head detectors, into the inference process. These detectors enable automatic initialization and aid recovery from transient tracking failures. We illustrate the method by automatically tracking people in multi-view imagery using a set of calibrated cameras and present quantitative evaluation using the HumanEva dataset.

We describe a complete system for animating realistic clothing on synthetic bodies of any shape and pose without manual intervention. The key component of the method is a model of clothing called DRAPE (DRessing Any PErson) that is learned from a physics-based simulation of clothing on bodies of different shapes and poses. The DRAPE model has the desirable property of "factoring" clothing deformations due to body shape from those due to pose variation. This factorization provides an approximation to the physical clothing deformation and greatly simplifies clothing synthesis. Given a parameterized model of the human body with known shape and pose parameters, we describe an algorithm that dresses the body with a garment that is customized to fit and possesses realistic wrinkles. DRAPE can be used to dress static bodies or animated sequences with a learned model of the cloth dynamics. Since the method is fully automated, it is appropriate for dressing large numbers of virtual characters of varying shape. The method is significantly more efficient than physical simulation.

The 3D shape of the human body is useful for applications in fitness, games and apparel. Accurate body scanners, however, are expensive, limiting the availability of 3D body models. We present a method for human shape reconstruction from noisy monocular image and range data using a single inexpensive commodity sensor. The approach combines low-resolution image silhouettes with coarse range data to estimate a parametric model of the body. Accurate 3D shape estimates are obtained by combining multiple monocular views of a person moving in front of the sensor. To cope with varying body pose, we use a SCAPE body model which factors 3D body shape and pose variations. This enables the estimation of a single consistent shape while allowing pose to vary. Additionally, we describe a novel method to minimize the distance between the projected 3D body contour and the image silhouette that uses analytic derivatives of the objective function. We propose a simple method to estimate standard body measurements from the recovered SCAPE model and show that the accuracy of our method is competitive with commercial body scanning systems costing orders of magnitude more.

The quantitative evaluation of optical flow algorithms by Barron et al. (1994) led to significant advances in performance. The challenges for optical flow algorithms today go beyond the datasets and evaluation methods proposed in that paper. Instead, they center on problems associated with complex natural scenes, including nonrigid motion, real sensor noise, and motion discontinuities. We propose a new set of benchmarks and evaluation methods for the next generation of optical flow algorithms. To that end, we contribute four types of data to test different aspects of optical flow algorithms: (1) sequences with nonrigid motion where the ground-truth flow is determined by tracking hidden fluorescent texture, (2) realistic synthetic sequences, (3) high frame-rate video used to study interpolation error, and (4) modified stereo sequences of static scenes. In addition to the average angular error used by Barron et al., we compute the absolute flow endpoint error, measures for frame interpolation error, improved statistics, and results at motion discontinuities and in textureless regions. In October 2007, we published the performance of several well-known methods on a preliminary version of our data to establish the current state of the art. We also made the data freely available on the web at http://vision.middlebury.edu/flow/ . Subsequently a number of researchers have uploaded their results to our website and published papers using the data. A significant improvement in performance has already been achieved. In this paper we analyze the results obtained to date and draw a large number of conclusions from them.

We define a new “contour person” model of the human body that has the expressive power of a detailed 3D model and the computational benefits of a simple 2D part-based model. The contour person (CP) model is learned from a 3D SCAPE model of the human body that captures natural shape and pose variations; the projected contours of this model, along with their segmentation into parts forms the training set. The CP model factors deformations of the body into three components: shape variation, viewpoint change and part rotation. This latter model also incorporates a learned non-rigid deformation model. The result is a 2D articulated model that is compact to represent, simple to compute with and more expressive than previous models. We demonstrate the value of such a model in 2D pose estimation and segmentation. Given an initial pose from a standard pictorial-structures method, we refine the pose and shape using an objective function that segments the scene into foreground and background regions. The result is a parametric, human-specific, image segmentation.

While the problem of tracking 3D human motion has been widely studied, most approaches have assumed that the person is isolated and not interacting with the environment. Environmental constraints, however, can greatly constrain and simplify the tracking problem. The most studied constraints involve gravity and contact with the ground plane. We go further to consider interaction with objects in the environment. In many cases, tracking rigid environmental objects is simpler than tracking high-dimensional human motion. When a human is in contact with objects in the world, their poses constrain the pose of body, essentially removing degrees of freedom. Thus what would appear to be a harder problem, combining object and human tracking, is actually simpler. We use a standard formulation of the body tracking problem but add an explicit model of contact with objects. We find that constraints from the world make it possible to track complex articulated human motion in 3D from a monocular camera.

The accuracy of optical flow estimation algorithms has been improving steadily as evidenced by results on the Middlebury optical flow benchmark. The typical formulation, however, has changed little since the work of Horn and Schunck. We attempt to uncover what has made recent advances possible through a thorough analysis of how the objective function, the optimization method, and modern implementation practices influence accuracy. We discover that “classical” flow formulations perform surprisingly well when combined with modern optimization and implementation techniques. Moreover, we find that while median filtering of intermediate flow fields during optimization is a key to recent performance gains, it leads to higher energy solutions. To understand the principles behind this phenomenon, we derive a new objective that formalizes the median filtering heuristic. This objective includes a nonlocal term that robustly integrates flow estimates over large spatial neighborhoods. By modifying this new term to include information about flow and image boundaries we develop a method that ranks at the top of the Middlebury benchmark.

Layered models are a powerful way of describing natural scenes containing smooth surfaces that may overlap and occlude each other. For image motion estimation, such models have a long history but have not achieved the wide use or accuracy of non-layered methods. We present a new probabilistic model of optical flow in layers that addresses many of the shortcomings of previous approaches. In particular, we define a probabilistic graphical model that explicitly captures: 1) occlusions and disocclusions; 2) depth ordering of the layers; 3) temporal consistency of the layer segmentation. Additionally the optical flow in each layer is modeled by a combination of a parametric model and a smooth deviation based on an MRF with a robust spatial prior; the resulting model allows roughness in layers. Finally, a key contribution is the formulation of the layers using an image-dependent hidden field prior based on recent models for static scene segmentation. The method achieves state-of-the-art results on the Middlebury benchmark and produces meaningful scene segmentations as well as detected occlusion regions.

We describe a solution to the challenging problem of estimating human body shape from a single photograph or painting. Our approach computes shape and pose parameters of a 3D human body model directly from monocular image cues and advances the state of the art in several directions. First, given a user-supplied estimate of the subject's height and a few clicked points on the body we estimate an initial 3D articulated body pose and shape. Second, using this initial guess we generate a tri-map of regions inside, outside and on the boundary of the human, which is used to segment the image using graph cuts. Third, we learn a low-dimensional linear model of human shape in which variations due to height are concentrated along a single dimension, enabling height-constrained estimation of body shape. Fourth, we formulate the problem of parametric human shape from shading. We estimate the body pose, shape and reflectance as well as the scene lighting that produces a synthesized body that robustly matches the image evidence. Quantitative experiments demonstrate how smooth shading provides powerful constraints on human shape. We further demonstrate a novel application in which we extract 3D human models from archival photographs and paintings.

Existing methods for recognition of object instances and categories based on quantized local features can perform poorly when local features exist on transparent surfaces, such as glass or plastic objects. There are characteristic patterns to the local appearance of transparent objects, but they may not be well captured by distances to individual examples or by a local pattern codebook obtained by vector quantization. The appearance of a transparent patch is determined in part by the refraction of a background pattern through a transparent medium: the energy from the background usually dominates the patch appearance. We model transparent local patch appearance using an additive model of latent factors: background factors due to scene content, and factors which capture a local edge energy distribution characteristic of the refraction. We implement our method using a novel LDA-SIFT formulation which performs LDA prior to any vector quantization step; we discover latent topics which are characteristic of particular transparent patches and quantize the SIFT space into transparent visual words according to the latent topic dimensions. No knowledge of the background scene is required at test time; we show examples recognizing transparent glasses in a domestic environment.

Much of the research on video-based human motion capture assumes the body shape is known a priori and is represented coarsely (e.g. using cylinders or superquadrics to model limbs). These body models stand in sharp contrast to the richly detailed 3D body models used by the graphics community. Here we propose a method for recovering such models directly from images. Specifically, we represent the body using a recently proposed triangulated mesh model called SCAPE which employs a low-dimensional, but detailed, parametric model of shape and pose-dependent deformations that is learned from a database of range scans of human bodies. Previous work showed that the parameters of the SCAPE model could be estimated from marker-based motion capture data. Here we go further to estimate the parameters directly from image data. We define a cost function between image observations and a hypothesized mesh and formulate the problem as optimization over the body shape and pose parameters using stochastic search. Our results show that such rich generative models enable the automatic recovery of detailed human shape and pose from images.

The quantitative evaluation of optical flow algorithms by Barron et al. led to significant advances in the performance of optical flow methods. The challenges for optical flow today go beyond the datasets and evaluation methods proposed in that paper and center on problems associated with nonrigid motion, real sensor noise, complex natural scenes, and motion discontinuities. Our goal is to establish a new set of benchmarks and evaluation methods for the next generation of optical flow algorithms. To that end, we contribute four types of data to test different aspects of optical flow algorithms: sequences with nonrigid motion where the ground-truth flow is determined by tracking hidden fluorescent texture; realistic synthetic sequences; high frame-rate video used to study interpolation error; and modified stereo sequences of static scenes. In addition to the average angular error used in Barron et al., we compute the absolute flow endpoint error, measures for frame interpolation error, improved statistics, and flow accuracy at motion boundaries and in textureless regions. We evaluate the performance of several well-known methods on this data to establish the current state of the art. Our database is freely available on the web together with scripts for scoring and publication of the results at http://vision.middlebury.edu/flow/.

Strong lighting is common in natural scenes yet is often viewed as a nuisance for object pose estimation and tracking. In human shape and pose estimation, cast shadows can be confused with foreground structure while self shadowing and shading variation on the body cause the appearance of the person to change with pose. Rather than attempt to minimize the effects of lighting and shadows, we show that strong lighting in a scene actually makes pose and shape estimation more robust. Additionally, by recovering multiple body poses we are able to automatically estimate the lighting in the scene and the albedo of the body. Our approach makes use of a detailed 3D body model, the parameters of which are directly recovered from image data. We provide a thorough exploration of human pose estimation under strong lighting conditions and show: 1. the estimation of the light source from cast shadows; 2. the estimation of the light source and the albedo of the body from multiple body poses; 3. that a point light and cast shadows on the ground plane can be treated as an additional "shadow camera" that improves pose and shape recovery, particularly in monocular scenes. Additionally we introduce the notion of albedo constancy which employs lighting normalized image data for matching. Our experiments with multiple subjects show that rather than causing problems, strong lighting improves human pose and shape estimation.

In contrast to traditional Markov random field (MRF) models, we develop a steerable random field (SRF) in which the field potentials are defined in terms of filter responses that are steered to the local image structure. In particular, we use the structure tensor to obtain derivative responses that are either aligned with, or orthogonal to, the predominant local image structure, and analyze the statistics of these steered filter responses in natural images. Clique potentials are defined over steered filter responses using a Gaussian scale mixture model and are learned from training data. The SRF model connects random field models with anisotropic regularization and provides a statistical motivation for the latter. We demonstrate that steering the random field to the local image structure improves image denoising and inpainting performance compared with traditional pairwise MRFs.

Estimation of three-dimensional articulated human pose and motion from images is a central problem in computer vision. Much of the previous work has been limited by the use of crude generative models of humans represented as articu- lated collections of simple parts such as cylinders. Automatic initialization of such models has proved difficult and most approaches assume that the size and shape of the body parts are known a priori. In this paper we propose a method for automatically recovering a detailed parametric model of non-rigid body shape and pose from monocular imagery. Specifically, we represent the body using a param- eterized triangulated mesh model that is learned from a database of human range scans. We demonstrate a discriminative method to directly recover the model pa- rameters from monocular images using a conditional mixture of kernel regressors. This predicted pose and shape are used to initialize a generative model for more detailed pose and shape estimation. The resulting approach allows fully automatic pose and shape recovery from monocular and multi-camera imagery. Experimen- tal results show that our method is capable of robustly recovering articulated pose, shape and biometric measurements (e.g. height, weight, etc.) in both calibrated and uncalibrated camera environments.

The detection and tracking of three-dimensional human body models has progressed rapidly but successful approaches typically rely on accurate foreground silhouettes obtained using background segmentation. There are many practical applications where such information is imprecise. Here we develop a new image likelihood function based on the visual appearance of the subject being tracked. We propose a robust, adaptive, appearance model based on the Wandering-Stable-Lost framework extended to the case of articulated body parts. The method models appearance using a mixture model that includes an adaptive template, frame-to-frame matching and an outlier process. We employ an annealed particle filtering algorithm for inference and take advantage of the 3D body model to predict selfocclusion and improve pose estimation accuracy. Quantitative tracking results are presented for a walking sequence with a 180 degree turn, captured with four synchronized and calibrated cameras and containing significant appearance changes and self-occlusion in each view.

In scenes containing specular objects, the image motion observed by a moving camera may be an intermixed combination of optical flow resulting from diffuse reflectance (diffuse flow) and specular reflection (specular flow). Here, with few assumptions, we formalize the notion of specular flow, show how it relates to the 3D structure of the world, and develop an algorithm for estimating scene structure from 2D image motion. Unlike previous work on isolated specular highlights we use two image frames and estimate the semi-dense flow arising from the specular reflections of textured scenes. We parametrically model the image motion of a quadratic surface patch viewed from a moving camera. The flow is modeled as a probabilistic mixture of diffuse and specular components and the 3D shape is recovered using an Expectation-Maximization algorithm. Rather than treating specular reflections as noise to be removed or ignored, we show that the specular flow provides additional constraints on scene geometry that improve estimation of 3D structure when compared with reconstruction from diffuse flow alone. We demonstrate this for a set of synthetic and real sequences of mixed specular-diffuse objects.

Part-based tree-structured models have been widely used for 2D articulated human pose-estimation. These approaches admit efficient inference algorithms while capturing the important kinematic constraints of the human body as a graphical model. These methods often fail however when multiple body parts fit the same image region resulting in global pose estimates that poorly explain the overall image evidence. Attempts to solve this problem have focused on the use of strong prior models that are limited to learned activities such as walking. We argue that the problem actually lies with the image observations and not with the prior. In particular, image evidence for each body part is estimated independently of other parts without regard to self-occlusion. To address this we introduce occlusion-sensitive local likelihoods that approximate the global image likelihood using per-pixel hidden binary variables that encode the occlusion relationships between parts. This occlusion reasoning introduces interactions between non-adjacent body parts creating loops in the underlying graphical model. We deal with this using an extension of an approximate belief propagation algorithm (PAMPAS). The algorithm recovers the real-valued 2D pose of the body in the presence of occlusions, does not require strong priors over body pose and does a quantitatively better job of explaining image evidence than previous methods.

We develop a Bayesian model of digitized archival films and use this for denoising, or more specifically de-graining, individual frames. In contrast to previous approaches our model uses a learned spatial prior and a unique likelihood term that models the physics that generates the image grain. The spatial prior is represented by a high-order Markov random field based on the recently proposed field-of-experts framework. We propose a new model of the image grain in archival films based on an inhomogeneous beta distribution in which the variance is a function of image luminance. We train this noise model for a particular film and perform de-graining using a diffusion method. Quantitative results show improved signal-to-noise ratio relative to the standard ad hoc Gaussian noise model.

Neural motor prostheses (NMPs) require the accurate decoding of motor cortical population activity for the control of an artificial motor system. Previous work on cortical decoding for NMPs has focused on the recovery of hand kinematics. Human NMPs however may require the control of computer cursors or robotic devices with very different physical and dynamical properties. Here we show that the firing rates of cells in the primary motor cortex of non-human primates can be used to control the parameters of an artificial physical system exhibiting realistic dynamics. The model represents 2D hand motion in terms of a point mass connected to a system of idealized springs. The nonlinear spring coefficients are estimated from the firing rates of neurons in the motor cortex. We evaluate linear and a nonlinear decoding algorithms using neural recordings from two monkeys performing two different tasks. We found that the decoded spring coefficients produced accurate hand trajectories compared with state-of-the-art methods for direct decoding of hand kinematics. Furthermore, using a physically-based system produced decoded movements that were more “natural” in that their frequency spectrum more closely matched that of natural hand movements.

Effective neural motor prostheses require a method for decoding neural activity representing desired movement. In particular, the accurate reconstruction of a continuous motion signal is necessary for the control of devices such as computer cursors, robots, or a patient's own paralyzed limbs. For such applications, we developed a real-time system that uses Bayesian inference techniques to estimate hand motion from the firing rates of multiple neurons. In this study, we used recordings that were previously made in the arm area of primary motor cortex in awake behaving monkeys using a chronically implanted multielectrode microarray. Bayesian inference involves computing the posterior probability of the hand motion conditioned on a sequence of observed firing rates; this is formulated in terms of the product of a likelihood and a prior. The likelihood term models the probability of firing rates given a particular hand motion. We found that a linear gaussian model could be used to approximate this likelihood and could be readily learned from a small amount of training data. The prior term defines a probabilistic model of hand kinematics and was also taken to be a linear gaussian model. Decoding was performed using a Kalman filter, which gives an efficient recursive method for Bayesian inference when the likelihood and prior are linear and gaussian. In off-line experiments, the Kalman filter reconstructions of hand trajectory were more accurate than previously reported results. The resulting decoding algorithm provides a principled probabilistic model of motor-cortical coding, decodes hand motion in real time, provides an estimate of uncertainty, and is straightforward to implement. Additionally the formulation unifies and extends previous models of neural coding while providing insights into the motor-cortical code.

We develop a framework for learning generic, expressive image priors that capture the statistics of natural scenes and can be used for a variety of machine vision tasks. The approach extends traditional Markov random field (MRF) models by learning potential functions over extended pixel neighborhoods. Field potentials are modeled using a Products-of-Experts framework that exploits nonlinear functions of many linear filter responses. In contrast to previous MRF approaches all parameters, including the linear filters themselves, are learned from training data. We demonstrate the capabilities of this Field of Experts model with two example applications, image denoising and image inpainting, which are implemented using a simple, approximate inference scheme. While the model is trained on a generic image database and is not tuned toward a specific application, we obtain results that compete with and even outperform specialized techniques.

In this work, we address the detection of vehicles in a video stream obtained from a moving airborne platform. We propose a Bayesian framework for estimating dense optical flow over time that explicitly estimates a persistent model of background appearance. The approach assumes that the scene can be described by background and occlusion layers, estimated within an expectation-maximization framework. The mathematical formulation of the paper is an extension of the work in (H. Yalcin et al., 2005) where motion and appearance models for foreground and background layers are estimated simultaneously in a Bayesian framework.

We develop a method for learning the spatial statistics of optical flow fields from a novel training database. Training flow fields are constructed using range images of natural scenes and 3D camera motions recovered from handheld and car-mounted video sequences. A detailed analysis of optical flow statistics in natural scenes is presented and machine learning methods are developed to learn a Markov random field model of optical flow. The prior probability of a flow field is formulated as a field-of-experts model that captures the higher order spatial statistics in overlapping patches and is trained using contrastive divergence. This new optical flow prior is compared with previous robust priors and is incorporated into a recent, accurate algorithm for dense optical flow computation. Experiments with natural and synthetic sequences illustrate how the learned optical flow prior quantitatively improves flow accuracy and how it captures the rich spatial structure found in natural scene motion.

Probabilistic modeling of correlated neural population firing activity is central to understanding the neural code and building practical decoding algorithms. No parametric models currently exist for modeling multivariate correlated neural data and the high dimensional nature of the data makes fully non-parametric methods impractical. To address these problems we propose an energy-based model in which the joint probability of neural activity is represented using learned functions of the 1D marginal histograms of the data. The parameters of the model are learned using contrastive divergence and an optimization procedure for finding appropriate marginal directions. We evaluate the method using real data recorded from a population of motor cortical neurons. In particular, we model the joint probability of population spiking times and 2D hand position and show that the likelihood of test data under our model is significantly higher than under other models. These results suggest that our model captures correlations in the firing activity. Our rich probabilistic model of neural population activity is a step towards both measurement of the importance of correlations in neural coding and improved decoding of population activity.

Many sensing techniques and image processing applications are characterized by noisy, or corrupted, image data. Anisotropic diffusion is a popular, and theoretically well understood, technique for denoising such images. Diffusion approaches however require the selection of an "edge stopping" function, the definition of which is typically ad hoc. We exploit and extend recent work on the statistics of natural images to define principled edge stopping functions for different types of imagery. We consider a variety of anisotropic diffusion schemes and note that they compute spatial derivatives at fixed scales from which we estimate the appropriate algorithm-specific image statistics. Going beyond traditional work on image statistics, we also model the statistics of the eigenvalues of the local structure tensor. Novel edge-stopping functions are derived from these image statistics giving a principled way of formulating anisotropic diffusion problems in which all edge-stopping parameters are learned from training data.

The detection and pose estimation of people in images and video is made challenging by the variability of human appearance, the complexity of natural scenes, and the high dimensionality of articulated body mod- els. To cope with these problems we represent the 3D human body as a graphical model in which the relationships between the body parts are represented by conditional probability distributions. We formulate the pose estimation problem as one of probabilistic inference over a graphi- cal model where the random variables correspond to the individual limb parameters (position and orientation). Because the limbs are described by 6-dimensional vectors encoding pose in 3-space, discretization is im- practical and the random variables in our model must be continuous- valued. To approximate belief propagation in such a graph we exploit a recently introduced generalization of the particle filter. This framework facilitates the automatic initialization of the body-model from low level cues and is robust to occlusion of body parts and scene clutter.

The direct neural control of external devices such as computer displays or prosthetic limbs requires the accurate decoding of neural activity rep- resenting continuous movement. We develop a real-time control system using the spiking activity of approximately 40 neurons recorded with an electrode array implanted in the arm area of primary motor cortex. In contrast to previous work, we develop a control-theoretic approach that explicitly models the motion of the hand and the probabilistic re- lationship between this motion and the mean firing rates of the cells in 70 bins. We focus on a realistic cursor control task in which the sub- ject must move a cursor to “hit” randomly placed targets on a computer monitor. Encoding and decoding of the neural data is achieved with a Kalman filter which has a number of advantages over previous linear filtering techniques. In particular, the Kalman filter reconstructions of hand trajectories in off-line experiments are more accurate than previ- ously reported results and the model provides insights into the nature of the neural coding of movement.  

We present a method for simultaneously learning linear models of multiple high dimensional data sets and the dependencies between them. For example, we learn asymmetrically coupled linear models for the faces of two different people and show how these models can be used to animate one face given a video sequence of the other. We pose the problem as a form of Asymmetric Coupled Component Analysis (ACCA) in which we simultaneously learn the subspaces for reducing the dimensionality of each dataset while coupling the parameters of the low dimensional representations. Additionally, a dynamic form of ACCA is proposed, that extends this work to model temporal dependencies in the data sets. To account for outliers and missing data, we formulate the problem in a statistically robust estimation framework. We review connections with previous work and illustrate the method with examples of synthesized dancing and the animation of facial avatars.

This paper describes a framework for learning probabilistic models of objects and scenes and for exploiting these models for tracking complex, deformable, or articulated objects in image sequences. We focus on the probabilistic tracking of people and learn models of how they appear and move in images. In particular we learn the likelihood of observing various spatial and temporal filter responses corresponding to edges, ridges, and motion differences given a model of the person. Similarly, we learn probability distributions over filter responses for general scenes that define a likelihood of observing the filter responses for arbitrary backgrounds. We then derive a probabilistic model for tracking that exploits the ratio between the likelihood that image pixels corresponding to the foreground (person) were generated by an actual person or by some unknown background. The paper extends previous work on learning image statistics and combines it with Bayesian tracking using particle filtering. By combining multiple image cues, and by using learned likelihood models, we demonstrate improved robustness and accuracy when tracking complex objects such as people in monocular image sequences with cluttered scene and a moving camera.

Statistical learning and probabilistic inference techniques are used to in- fer the hand position of a subject from multi-electrode recordings of neu- ral activity in motor cortex. First, an array of electrodes provides train- ing data of neural firing conditioned on hand kinematics. We learn a non- parametric representation of this firing activity using a Bayesian model and rigorously compare it with previous models using cross-validation. Second, we infer a posterior probability distribution over hand motion conditioned on a sequence of neural test data using Bayesian inference. The learned firing models of multiple cells are used to define a non- Gaussian likelihood term which is combined with a prior probability for the kinematics. A particle filtering method is used to represent, update, and propagate the posterior distribution over time. The approach is com- pared with traditional linear filtering methods; the results suggest that it may be appropriate for neural prosthetic applications.

This paper describes a framework for constructing a linear subspace model of image appearance for complex articulated 3D figures such as humans and other animals. A commercial motion capture system provides 3D data that is aligned with images of subjects performing various activities. Portions of a limb's image appearance are seen from multiple views and for multiple subjects. From these partial views, weighted principal component analysis is used to construct a linear subspace representation of the "unwrapped" image appearance of each limb. The linear subspaces provide a generative model of the object appearance that is exploited in a Bayesian particle filtering tracking system. Results of tracking single limbs and walking humans are presented.

We present methods for learning and tracking human motion in video. We estimate a statistical model of typical activities from a large set of 3D periodic human motion data by segmenting these data automatically into "cycles". Then the mean and the princi(cid:173) pal components of the cycles are computed using a new algorithm that accounts for missing information and enforces smooth tran(cid:173) sitions between cycles. The learned temporal model provides a prior probability distribution over human motions that can be used in a Bayesian framework for tracking human subjects in complex monocular video sequences and recovering their 3D motion.

A spatio-temporal representation for complex optical flow events is developed that generalizes traditional parameterized motion models (e.g. affine). These generative spatio-temporal models may be non-linear or stochastic and are event-specific in that they characterize a particular type of object motion (e.g. sitting or walking). Within a Bayesian framework we seek the appropriate model, phase, rate, spatial position, and scale to account for the image variation. The posterior distribution over this parameter space conditioned on image measurements is typically non-Gaussian. The distribution is represented using factored sampling and is predicted and updated over time using the condensation algorithm. The resulting framework automatically detects, localizes, and recognizes motion events.

We propose a Bayesian framework for representing and recognizing local image motion in terms of two primitive models: translation and motion discontinuity. Motion discontinuities are represented using a nonlinear generative model that explicitly encodes the orientation of the boundary, the velocities on either side, the motion of the occluding edge over time, and the appearance/disappearance of pixels at the boundary. We represent the posterior distribution over the model parameters given the image data using discrete samples. This distribution is propagated over time using the Condensation algorithm. To efficiently represent such a high-dimensional space we initialize samples using the responses of a low-level motion discontinuity detector.

The estimation and detection of occlusion boundaries and moving bars are important and challenging problems in image sequence analysis. Here, we model such motion features as linear combinations of steerable basis flow fields. These models constrain the interpretation of image motion, and are used in the same way as translational or affine motion models. We estimate the subspace coefficients of the motion feature models directly from spatiotemporal image derivatives using a robust regression method. From the subspace coefficients we detect the presence of a motion feature and solve for the orientation of the feature and the relative velocities of the surfaces. Our method does not require the prior computation of optical flow and recovers accurate estimates of orientation and velocity.

Image "appearance" may change over time due to a variety of causes such as: 1) object or camera motion; 2) generic photometric events including variations in illumination (e.g. shadows) and specular reflections; and 3) "iconic changes" which are specific to the objects being viewed and include complex occlusion events and changes in the material properties of the objects. We propose a general framework for representing and recovering these "appearance changes" in an image sequence as a "mixture" of different causes. The approach generalizes previous work on optical flow to provide a richer description of image events and more reliable estimates of image motion.

A framework for modeling and recognition of temporal activities is proposed. The modeling of sets of exemplar activities is achieved by parameterizing their representation in the form of principal components. Recognition of spatio-temporal variants of modeled activities is achieved by parameterizing the search in the space of admissible transformations that the activities can undergo. Experiments on recognition of articulated and deformable object motion from image motion parameters are presented.

This paper presents an automatic system for analyzing and annotating video sequences of technical talks. Our method uses a robust motion estimation technique to detect key frames and segment the video sequence into subsequences containing a single overhead slide. The subsequences are stabilized to remove motion that occurs when the speaker adjusts their slides. Any changes remaining between frames in the stabilized sequences may be due to speaker gestures such as pointing or writing, and we use active contours to automatically track these potential gestures. Given the constrained domain, we define a simple set of actions that can be recognized based on the active contour shape and motion. The recognized actions provide an annotation of the sequence that can be used to access a condensed version of the talk from a Web page.

Relations between anisotropic diffusion and robust statistics are described in this paper. Specifically, we show that anisotropic diffusion can be seen as a robust estimation procedure that estimates a piecewise smooth image from a noisy input image. The "edge-stopping" function in the anisotropic diffusion equation is closely related to the error norm and influence function in the robust estimation framework. This connection leads to a new "edge-stopping" function based on Tukey's biweight robust estimator that preserves sharper boundaries than previous formulations and improves the automatic stopping of the diffusion. The robust statistical interpretation also provides a means for detecting the boundaries (edges) between the piecewise smooth regions in an image that has been smoothed with anisotropic diffusion. Additionally, we derive a relationship between anisotropic diffusion and regularization with line processes. Adding constraints on the spatial organization of the line processes allows us to develop new anisotropic diffusion equations that result in a qualitative improvement in the continuity of edges.

A framework for learning parameterized models of optical flow from image sequences is presented. A class of motions is represented by a set of orthogonal basis flow fields that are computed from a training set using principal component analysis. Many complex image motions can be represented by a linear combination of a small number of these basis flows. The learned motion models may be used for optical flow estimation and for model-based recognition. For optical flow estimation we describe a robust, multi-resolution scheme for directly computing the parameters of the learned flow models from image derivatives. As examples we consider learning motion discontinuities, non-rigid motion of human mouths, and articulated human motion.

We present an automatic system for analyzing and annotating video sequences of technical talks. Our method uses a robust motion estimation technique to detect key frames and segment the video sequence into subsequences containing a single overhead slide. The subsequences are stabilized to remove motion that occurs when the speaker adjusts their slides. Any changes remaining between frames in the stabilized sequences may be due to speaker gestures such as pointing or writing and we use active contours to automatically track these potential gestures. Given the constrained domain we define a simple "vocabulary" of actions which can easily be recognized based on the active contour shape and motion. The recognized actions provide a rich annotation of the sequence that can be used to access a condensed version of the talk from a web page.

Relations between anisotropic diffusion and robust statistics are described. We show that anisotropic diffusion can be seen as a robust estimation procedure that estimates a piecewise smooth image from a noisy input image. The "edge-stopping" function in the anisotropic diffusion equation is closely related to the error norm and influence function in the robust estimation framework. This connection leads to a new "edge-stopping" function based on Tukey's biweight robust estimator, that preserves sharper boundaries than previous formulations and improves the automatic stopping of the diffusion. The robust statistical interpretation also provides a means for detecting the boundaries (edges) between the piecewise smooth regions in the image. We extend the framework to vector-valued images and show applications to robust image sharpening.

This paper describes a new method for estimating optical flow that strikes a balance between the flexibility of local dense computations and the robustness and accuracy of global parameterized flow models. An affine model of image motion is used within local image patches while a spatial smoothness constraint on the affine flow parameters of neighboring patches enforces continuity of the motion. We refer to this as a "Skin and Bones" model in which the affine patches can be thought of as rigid "bones" connected by a flexible "skin". Since local image patches may contain multiple motions we use a layered representation for the affine bones. To regularize this layered motion representation we develop a new framework for regularization with transparency.

We extend the work of Black and Yacoob (1995) on the tracking and recognition of human facial expressions using parametrized models of optical flow to deal with the articulated motion of human limbs. We define a "card-board person model" in which a person's limbs are represented by a set of connected planar patches. The parametrized image motion of these patches in constrained to enforce articulated motion and is solved for directly using a robust estimation technique. The recovered motion parameters provide a rich and concise description of the activity that can be used for recognition. We propose a method for performing view-based recognition of human activities from the optical flow parameters that extends previous methods to cope with the cyclical nature of human motion. We illustrate the method with examples of tracking human legs of long image sequences.

This paper presents a new model for estimating optical flow based on the motion of planar regions plus local deformations. The approach exploits brightness information to organize and constrain the interpretation of the motion by using segmented regions of piecewise smooth brightness to hypothesize planar regions in the scene. Parametric flow models are estimated in these regions in a two step process which first computes a coarse fit and then estimates the appropriate parametrization of the motion of the region. The initial fit is refined using a generalization of the standard area-based regression approaches. Since the assumption of planarity is likely to be violated, we allow local deformations from the planar assumption in the same spirit as physically-based approaches which model shape using coarse parametric models plus local deformations. This parametric plus deformation model exploits the strong constraints of parametric approaches while retaining the adaptive nature of regularization approaches. Experimental results on a variety of images model produces accurate flow estimates while the incorporation of brightness segmentation boundaries.

This paper explores the use of local parametrized models of image motion for recovering and recognizing the non-rigid and articulated motion of human faces. Parametric flow models (for example affine) are popular for estimating motion in rigid scenes. We observe that within local regions in space and time, such models not only accurately model non-rigid facial motions but also provide a concise description of the motion in terms of a small number of parameters. These parameters are intuitively related to the motion of facial features during facial expressions and we show how expressions such as anger, happiness, surprise, fear, disgust and sadness can be recognized from the local parametric motions in the presence of significant head motion. The motion tracking and expression recognition approach performs with high accuracy in extensive laboratory experiments involving 40 subjects as well as in television and movie sequences.>

This paper unifies "line-process" approaches for regularization with discontinuities and robust estimation techniques. We generalize the notion of a "line process" to that of an analog "outlier process" and show that a problem formulated in terms of outlier processes can be viewed in terms of robust statistics. We also characterize a class of robust statistical problems for which an equivalent outlier-process formulation exists and give a straightforward method for converting a robust estimation problem into an outlier-process formulation. This outlier-processes approach provides a general framework which subsumes the traditional line-process approaches as well as a wide class of robust estimation problems. Examples in image reconstruction and optical flow are used to illustrate the approach.>

The computation of optical flow relies on merging information available over an image patch to form an estimate of 2-D image velocity at a point. This merging process raises many issues. These include the treatment of outliers in component velocity measurements and the modeling of multiple motions within a patch which arise from occlusion boundaries or transparency. A new approach for dealing with these issues is presented. It is based on the use of a probabilistic mixture model to explicitly represent multiple motions within a patch. A simple extension of the EM-algorithm is used to compute a maximum likelihood estimate for the various motion parameters. Preliminary experiments indicate that this approach is computationally efficient, and that it can provide robust estimates of the optical flow values in the presence of outliers and multiple motions.>

The authors consider the problem of robustly estimating optical flow from a pair of images using a new framework based on robust estimation which addresses violations of the brightness constancy and spatial smoothness assumptions. They also show the relationship between the robust estimation framework and line-process approaches for coping with spatial discontinuities. In doing so, the notion of a line process is generalized to that of an outlier process that can account for violations in both the brightness and smoothness assumptions. A graduated non-convexity algorithm is presented for recovering optical flow and motion discontinuities. The performance of the robust formulation is demonstrated on both synthetic data and natural images.>

A novel approach to incrementally estimating visual motion over a sequence of images is presented. The authors start by formulating constraints on image motion to account for the possibility of multiple motions. This is achieved by exploiting the notions of weak continuity and robust statistics in the formulation of a minimization problem. The resulting objective function is non-convex. Traditional stochastic relaxation techniques for minimizing such functions prove inappropriate for the task. A highly parallel incremental stochastic minimization algorithm is presented which has a number of advantages over previous approaches. The incremental nature of the scheme makes it dynamic and permits the detection of occlusion and disocclusion boundaries.>

A model is proposed for the incremental estimation of visual motion fields from image sequences. The authors' model exploits three standard constraints on image motion within an optimization framework: (1) data conservation-the intensity structure of a surface patch changes gradually over time; (2) spatial coherence-neighboring points have similar motions; and (3) temporal coherence-the image velocity of a surface patch changes gradually. The authors' formulation takes into account the possibility of multiple motions at a particular location. They present an incremental scheme for the minimization of the objective function, based on simulated annealing. All computations are parallel, local, and incremental, and occlusion and disocclusion boundaries are estimated.>