We introduce a novel representation for efficient classical rendering of photorealistic 3D face avatars. Leveraging recent advances in radiance fields anchored to parametric face models, our approach achieves controllable volumetric rendering of complex facial features, including hair, skin, and eyes. At enrollment time, we learn a set of radiance manifolds in 3D space to extract an explicit layered mesh, along with appearance and warp textures. During deployment, this allows us to control and animate the face through simple linear blending and alpha compositing of textures over a static mesh. This explicit representation also enables the generated avatar to be efficiently streamed online and then rendered using classical mesh and shader-based rendering on legacy graphics platforms, eliminating the need for any custom engineering or integration. https://syntec-research.github.io/LegacyAvatars/

We introduce GarmentCrafter, a new approach to enable non-professional users to create and modify 3D garments from a single-view image. While recent advances in image generation have facilitated$2 D$garment design, creating and editing 3D garments remains challenging for nonprofessional users. Existing methods for single-view 3D reconstruction often rely on pre-trained generative models to hallucinate novel views conditioning on the reference image and camera pose, yet they lack cross-view consistency, failing to capture the internal relationships across different views. In this paper, we tackle this challenge through progressive depth prediction and image warping to approximate novel views. Subsequently, we train a multi-view diffusion model to complete occluded and unknown clothing regions, informed by the evolving camera pose. By jointly inferring RGB and depth, GarmentCrafter enforces inter-view coherence and reconstructs precise geometries and fine details. Extensive experiments demonstrate that our method achieves superior visual fidelity and inter-view coherence compared to state-of-the-art single-view 3D garment reconstruction methods. Our code will be publicly available.

Driving a high-quality and photorealistic full-body virtual human from a few RGB cameras is a challenging problem that has become increasingly relevant with emerging virtual reality technologies. A promising solution to democratize such technology would be a generalizable method that takes sparse multi-view images of any person and then generates photoreal free-view renderings of them. However, the state-of-the-art approaches are not scalable to very large datasets and, thus, lack diversity and photorealism. To address this problem, we propose GIGA, a novel, generalizable full-body model for rendering photoreal humans in free viewpoint, driven by a single-view or sparse multi-view video. Notably, GIGA can scale training to a few thousand subjects while maintaining high photorealism and synthesizing dynamic appearance. At the core, we introduce a MultiHeadUNet architecture, which takes an approximate RGB texture accumulated from a single or multiple sparse views and predicts 3D Gaussian primitives represented as 2D texels on top of a human body mesh. At test time, our method performs novel view synthesis of a virtual 3D Gaussian-based human from 1 to 4 input views and a tracked body template for unseen identities. Our method excels over prior works by a significant margin in terms of identity generalization capability and photorealism.

Mixed Reality (MR) headsets promise a future of immersive telepresence where virtual humans blend indistinguishably into their real or virtual surroundings. Achieving this vision requires a method capable of capturing a user's motion, estimating their appearance under novel lighting, and understanding the surrounding environment - all from the constrained viewpoint of a head-mounted display (HMD). Existing approaches treat these as isolated problems: they either focus on driving avatars with baked-in lighting or rely on complex studio setups for relighting. In this paper, we present EgoRelight , a holistic framework for egocentric telepresence that simultaneously captures full-body human performance, synthesizes photorealistic and relightable appearance, and estimates high dynamic range (HDR) environment maps from a single HMD. First, to ensure accurate motion and surface reconstruction, we propose an egocentric perception module that leverages stereo down-facing cameras to extract dense depth maps, which serve as robust geometric control signals to drive a mesh-based avatar. Second, we introduce a novel neural appearance model that learns to synthesize view-dependent specular and view-independent diffuse shading separately. By employing a specialized ray-sampling strategy, our model generalizes to unseen illumination without relying on restrictive analytical BRDF priors. Third, we enable seamless avatar integration into the physical world via a test-time inverse rendering process, which recovers an HDR environment map by matching the pre-trained avatar's appearance to the live egocentric camera observations. We demonstrate the efficiency of our system through a social telepresence application, where remote users are coherently relit according to their local physical environment. Extensive experiments show that our individual components and the integrated system significantly outperform the best combinations of state-of-the-art baselines in both geometric accuracy and rendering as well as relighting fidelity. Further details can be found on our project page 1 .

This work focuses on tracking and understanding human motion using consumer wearable devices, such as VR/AR headsets, smart glasses, cellphones, and smartwatches. These devices provide diverse, multi-modal sensor inputs, including egocentric images, and 1-3 sparse IMU sensors in varied combinations. Motion descriptions can also accompany these signals. The diverse input modalities and their intermittent availability pose challenges for consistent motion capture and understanding. In this work, we present Ego4o (o for omni), a new framework for simultaneous human motion capture and understanding from multi-modal egocentric inputs. This method maintains performance with partial inputs while achieving better results when multiple modalities are combined. First, the IMU sensor inputs, the optional egocentric image, and text description of human motion are encoded into the latent space of a motion VQ-VAE. Next, the latent vectors are sent to the VQ-VAE decoder and optimized to track human motion. When motion descriptions are unavailable, the latent vectors can be input into a multi-modal LLM to generate human motion descriptions, which can further enhance motion capture accuracy. Quantitative and qualitative evaluations demonstrate the effectiveness of our method in predicting accurate human motion and high-quality motion descriptions. Project page: https://jianwang-mpi.github.io/ego4o.

Egocentric motion capture with a head-mounted body-facing stereo camera is crucial for VR and AR applications but presents significant challenges such as heavy occlusions and limited annotated real-world data. Existing methods rely on synthetic pretraining and struggle to generate smooth and accurate predictions in real-world settings, particularly for lower limbs. Our work addresses these limitations by introducing a lightweight VR-based data collection setup with on-board, real-time 6D pose tracking. Using this setup, we collected the most extensive real-world dataset for ego-facing ego-mounted cameras to date in size and motion variability. Effectively integrating this multimodal input –device pose and camera feeds –is challenging due to the differing characteristics of each data source. To address this, we propose FRAME, a simple yet effective architecture that combines device pose and camera feeds for state-of-the-art body pose prediction through geometrically sound multimodal integration and can run at 300 FPS on modern hardware. Lastly, we showcase a novel training strategy to enhance the model’s generalization capabilities. Our approach exploits the problem’s geometric properties, yielding high-quality motion capture free from common artifacts in prior work. Qualitative and quantitative evaluations, along with extensive comparisons, demonstrate the effectiveness of our method. Data, code, and CAD designs will be available at vcai.mpi-inf.mpg.de/projects/FRAME.

We present BimArt, a novel generative approach for synthesizing 3D bimanual hand interactions with articulated objects. Unlike prior works, we do not rely on a reference grasp, a coarse hand trajectory, or separate modes for grasping and articulating. To achieve this, we first generate distance-based contact maps conditioned on the object trajectory with an articulation-aware feature representation, revealing rich bimanual patterns for manipulation. The learned contact prior is then used to guide our hand motion generator, producing diverse and realistic bimanual motions for object movement and articulation. Our work offers key insights into feature representation and contact prior for articulated objects, demonstrating their effectiveness in taming the complex, high-dimensional space of bimanual hand-object interactions. Through comprehensive quantitative experiments, we demonstrate a clear step towards simplified and high-quality hand-object animations that surpass the state of the art in motion quality and diversity. Project page: https://vcai.mpi-inf.mpg.de/projects/bimart/.

We present GroomLight, a novel method for relightable hair appearance modeling from multi-view images. Existing hair capture methods struggle to balance photorealistic rendering with relighting capabilities. Analytical material models, while physically grounded, often fail to fully capture appearance details. Conversely, neural rendering approaches excel at view synthesis but generalize poorly to novel lighting conditions. GroomLight addresses this challenge by combining the strengths of both paradigms. It employs an extended hair BSDF model to capture primary light transport and a light-aware residual model to reconstruct the remaining details. We further propose a hybrid inverse rendering pipeline to optimize both components, enabling high-fidelity relighting, view synthesis, and material editing. Extensive evaluations on real-world hair data demonstrate state-of-the-art performance of our method. Our project website is at: https://syntec-research.github.io/GroomLight.

Current personalized neural head avatars face a trade-off: lightweight models lack detail and realism, while high-quality, animatable avatars require significant computational resources, making them unsuitable for commodity devices. To address this gap, we introduce Gaussian Eigen Models (GEM), which provide high-quality, lightweight, and easily controllable head avatars. GEM utilizes 3D Gaussian primitives for representing the appearance combined with Gaussian splatting for rendering. Building on the success of mesh-based 3D morphable face models (3DMM), we define GEM as an ensemble of linear eigen-bases for representing the head appearance of a specific subject. In particular, we construct linear bases to represent the position, scale, rotation, and opacity of the 3D Gaussians. This allows us to efficiently generate Gaussian primitives of a specific head shape by a linear combination of the basis vectors, only requiring a low-dimensional parameter vector that contains the respective coefficients. We propose to construct these linear bases (GEM) by distilling high-quality compute-intense CNN-based Gaussian avatar models that can generate expression-dependent appearance changes like wrinkles. These high-quality models are trained on multi-view videos of a subject and are dis tilled using a series of principle component analyses.Once we have obtained the bases that represent the animatable appearance space of a specific human, we learn a regressor that takes a single RGB image as input and predicts the low-dimensional parameter vector that corresponds to the shown facial expression. We demonstrate that this regressor can be trained such that it effectively supports self- and cross-person reenactment from monocular videos without requiring prior mesh-based tracking. In a series of experiments, we compare GEM’s self-reenactment and cross-person reenactment results to state-of-the-art 3D avatar methods, demonstrating GEM’s higher visual quality and better generalization to new expressions. As our distilled linear model is highly efficient in generating novel animation states, we also show a real-time demo of GEMs driven by monocular webcam videos. The code and model will be released for research purposes.

We present SynShot, a novel method for the few-shot inversion of a drivable head avatar based on a synthetic prior. We tackle three major challenges. First, training a controllable 3D generative network requires a large number of diverse sequences, for which pairs of images and high-quality tracked meshes are not always available. Second, the use of real data is strictly regulated (e.g., under the General Data Protection Regulation, which mandates frequent deletion of models and data to accommodate a situation when participant’s consent is withdrawn). Synthetic data, free from these constraints, is an appealing alternative. Third, state-of-the-art monocular avatar models struggle to generalize to new views and expressions, lacking a strong prior and often overfitting to a specific viewpoint distribution. Inspired by machine learning models trained solely on synthetic data, we propose a method that learns a prior model from a large dataset of synthetic heads with diverse identities, expressions, and viewpoints. With few input images, SynShot fine-tunes the pretrained synthetic prior to bridge the domain gap, modeling a photorealistic head avatar that generalizes to novel expressions and viewpoints. We model the head avatar using 3D Gaussian splatting and a convolutional encoder-decoder that outputs Gaussian parameters in UV texture space. To account for the different modeling complexities over parts of the head (e.g., skin vs hair), we embed the prior with explicit control for upsampling the number of per-part primitives. Compared to SOTA monocular and GAN-based methods, SynShot significantly improves novel view and expression synthesis.

Digital humans are emerging as autonomous agents in multiparty interactions, yet existing evaluation metrics largely ignore contextual coordination dynamics. We introduce a unified, intervention-driven framework for objective assessment of multiparty social behaviour in skeletal motion data, spanning three complementary dimensions: (1) synchrony via Cross-Recurrence Quantification Analysis, (2) temporal alignment via Multiscale Empirical Mode Decompositionbased Beat Consistency, and (3) structural similarity via Soft Dynamic Time Warping. We validate metric sensitivity through three theory-driven perturbations -- gesture kinematic dampening, uniform speech-gesture delays, and prosodic pitch-variance reduction-applied to $\approx 145$ 30-second thin slices of group interactions from the DnD dataset. Mixed-effects analyses reveal predictable, joint-independent shifts: dampening increases CRQA determinism and reduces beat consistency, delays weaken cross-participant coupling, and pitch flattening elevates F0 Soft-DTW costs. A complementary perception study ($N=27$) compares judgments of full-video and skeleton-only renderings to quantify representation effects. Our three measures deliver orthogonal insights into spatial structure, timing alignment, and behavioural variability. Thereby forming a robust toolkit for evaluating and refining socially intelligent agents. Code available on \href{https://github.com/tapri-lab/gig-interveners}{GitHub}.

Rendering novel, relit views of a human head, given a monocular portrait image as input, is an inherently underconstrained problem. The traditional graphics solution is to explicitly decompose the input image into geometry, material and lighting via differentiable rendering; but this is constrained by the multiple assumptions and approximations of the underlying models and parameterizations of these scene components. We propose 3DPR, an image-based relighting model that leverages generative priors learnt from multi-view One-Light-at-A-Time (OLAT) images captured in a light stage. We introduce a new diverse and large-scale multi-view 4K OLAT dataset of 139 subjects to learn a high-quality prior over the distribution of high-frequency face reflectance. We leverage the latent space of a pre-trained generative head model that provides a rich prior over face geometry learnt from in-the-wild image datasets. The input portrait is first embedded in the latent manifold of such a model through an encoder-based inversion process. Then a novel triplane-based reflectance network trained on our lightstage data is used to synthesize high-fidelity OLAT images to enable image-based relighting. Our reflectance network operates in the latent space of the generative head model, crucially enabling a relatively small number of lightstage images to train the reflectance model. Combining the generated OLATs according to a given HDRI environment maps yields physically accurate environmental relighting results. Through quantitative and qualitative evaluations, we demonstrate that 3DPR outperforms previous methods, particularly in preserving identity and in capturing lighting effects such as specularities, self-shadows, and subsurface scattering.

Inverse rendering has emerged as a standard tool to reconstruct the parameters of appearance models from images (e.g., textured BSDFs). In this work, we present several novel contributions motivated by the practical challenges of recovering high-resolution surface appearance textures, including spatially-varying subsurface scattering parameters. First, we propose Laplacian mipmapping , which combines differentiable mipmapping and a Laplacian pyramid representation into an effective preconditioner. This seemingly simple technique significantly improves the quality of recovered surface textures on a set of challenging inverse rendering problems. Our method automatically adapts to the render and texture resolutions, only incurs moderate computational cost and achieves better quality than prior work while using fewer hyperparameters. Second, we introduce a specialized gradient computation algorithm for textured, path-traced subsurface scattering, which facilitates faithful reconstruction of translucent materials. By using path tracing, we enable the recovery of complex appearance while avoiding the approximations of the previously used diffusion dipole methods. Third, we demonstrate the application of both these techniques to reconstructing the textured appearance of human faces from sparse captures. Our method recovers high-quality relightable appearance parameters that are compatible with current production renderers.

The physical properties of an object, such as mass, significantly affect how we manipulate it with our hands. Surprisingly, this aspect has so far been neglected in prior work on 3D motion synthesis. To improve the naturalness of the synthesized 3D hand-object motions, this work proposes MACS–the first MAss Conditioned 3D hand and object motion Synthesis approach. Our approach is based on cascaded diffusion models and generates interactions that plausibly adjust based on the object’s mass and interaction type. MACS also accepts a manually drawn 3D object trajectory as input and synthesizes the natural 3D hand motions conditioned by the object’s mass. This flexibility enables MACS to be used for various downstream applications, such as generating synthetic training data for ML tasks, fast animation of hands for graphics workflows, and generating character interactions for computer games. We show experimentally that a small-scale dataset is sufficient for MACS to reasonably generalize across interpolated and extrapolated object masses unseen during the training. Furthermore, MACS shows moderate generalization to unseen objects, thanks to the mass-conditioned contact labels generated by our surface contact synthesis model ConNet. Our comprehensive user study confirms that the synthesized 3D hand-object interactions are highly plausible and realistic.

We propose Diffusion Noise Optimization (DNO), a new method that effectively leverages existing motion diffusion models as motion priors for a wide range of motion-related tasks. Instead of training a task-specific diffusion model for each new task, DNO operates by optimizing the diffusion latent noise of an existing pre-trained text-to-motion model. Given the corresponding latent noise of a human motion, it propagates the gradient from the target criteria defined on the motion space through the whole denoising process to update the diffusion latent noise. As a result, DNO supports any use cases where criteria can be defined as a function of motion. In particular, we show that, for motion editing and control, DNO outperforms existing meth-ods in both achieving the objective and preserving the motion content. DNO accommodates a diverse range of editing modes, including changing trajectory, pose, joint lo-cations, or avoiding newly added obstacles. In addition, DNO is effective in motion denoising and completion, pro-ducing smooth and realistic motion from noisy and partial inputs. DNO achieves these results at inference time with-out the need for model retraining, offering great versatility for any defined reward or loss function on the motion rep-resentation.

In this work, we explore egocentric whole-body motion capture using a single fisheye camera, which simultane-ously estimates human body and hand motion. This task presents significant challenges due to three factors: the lack of high-quality datasets, fisheye camera distortion, and hu-man body self-occlusion. To address these challenges, we propose a novel approach that leverages Fisheye ViT to ex-tract fisheye image features, which are subsequently con-verted into pixel-aligned 3D heatmap representations for 3D human body pose prediction. For hand tracking, we incorporate dedicated hand detection and hand pose esti-mation networks for regressing 3D hand poses. Finally, we develop a diffusion-based whole-body motion prior model to refine the estimated whole-body motion while accounting for joint uncertainties. To train these networks, we col-lect a large synthetic dataset, Ego WholeBody, comprising 840,000 high-quality egocentric images captured across a diverse range of whole-body motion sequences. Quantitative and qualitative evaluations demonstrate the effective-ness of our method in producing high-quality whole-body motion estimates from a single egocentric camera.

Volumetric modeling and neural radiance field representations have revolutionized 3D face capture and photorealistic novel view synthesis. However, these methods often require hundreds of multi-view input images and are thus inapplicable to cases with less than a handful of inputs. We present a novel volumetric prior on human faces that allows for high-fidelity expressive face modeling from as few as three input views captured in the wild. Our key insight is that an implicit prior trained on synthetic data alone can generalize to extremely challenging real-world identities and expressions and render novel views with fine idiosyncratic details like wrinkles and eyelashes. We leverage a 3D Morphable Face Model to synthesize a large training set, rendering each identity with different expressions, hair, clothing, and other assets. We then train a conditional Neural Radiance Field prior on this synthetic dataset and, at inference time, fine-tune the model on a very sparse set of real images of a single subject. On average, the fine-tuning requires only three inputs to cross the synthetic-to-real domain gap. The resulting personalized 3D model reconstructs strong idiosyncratic facial expressions and outperforms the state-of-the-art in high-quality novel view synthesis of faces from sparse inputs in terms of perceptual and photo-metric quality.

Abstract Nowadays, it is possible to scan faces and automatically register them with high quality. However, the resulting face meshes often need further processing: we need tostabilizethem to remove unwanted head movement. Stabilization is important for tasks like game development or movie making which require facial expressions to be cleanly separated from rigid head motion. Since manual stabilization is labor‐intensive, there have been attempts to automate it. However, previous methods remain impractical: they either still require some manual input, produce imprecise alignments, rely on dubious heuristics and slow optimization, or assume a temporally ordered input. Instead, we present a new learning‐based approach that is simple and fully automatic. We treat stabilization as a regression problem: given two face meshes, our network directly predicts the rigid transform between them that brings their skulls into alignment. We generate synthetic training data using a 3D Morphable Model (3DMM), exploiting the fact that 3DMM parameters separate skull motion from facial skin motion. Through extensive experiments we show that our approach outperforms the state‐of‐the‐art both quantitatively and qualitatively on the tasks of stabilizing discrete sets of facial expressions as well as dynamic facial performances. Furthermore, we provide an ablation study detailing the design choices and best practices to help others adopt our approach for their own uses.

Abstract High‐resolution texture maps are essential to render photoreal digital humans for visual effects or to generate data for machine learning. The acquisition of high resolution assets at scale is cumbersome, it involves enrolling a large number of human subjects, using expensive multi‐view camera setups, and significant manual artistic effort to align the textures. To alleviate these problems, we introduce GANtlitz (A play on the german noun Antlitz, meaning face), a generative model that can synthesize multi‐modal ultra‐high‐resolution face appearance maps for novel identities. Our method solves three distinct challenges: 1) unavailability of a very large data corpus generally required for training generative models, 2) memory and computational limitations of training a GAN at ultra‐high resolutions, and 3) consistency of appearance features such as skin color, pores and wrinkles in high‐resolution textures across different modalities. We introduce dual‐style blocks, an extension to the style blocks of the StyleGAN2 architecture, which improve multi‐modal synthesis. Our patch‐based architecture is trained only on image patches obtained from a small set of face textures (<100) and yet allows us to generate seamless appearance maps of novel identities at 6k × 4k resolution. Extensive qualitative and quantitative evaluations and baseline comparisons show the efficacy of our proposed system. (see https://www.acm.org/publications/class-2012 )

Abstract Eye gaze and expressions are crucial non‐verbal signals in face‐to‐face communication. Visual effects and telepresence demand significant improvements in personalized tracking, animation, and synthesis of the eye region to achieve true immersion. Morphable face models, in combination with coordinate‐based neural volumetric representations, show promise in solving the difficult problem of reconstructing intricate geometry (eyelashes) and synthesizing photorealistic appearance variations (wrinkles and specularities) of eye performances. We propose a novel hybrid representation ‐ ShellNeRF ‐ that builds a discretized volume around a 3DMM face mesh using concentric surfaces to model the deformable ‘periocular’ region. We define a canonical space using the UV layout of the shells that constrains the space of dense correspondence search. Combined with an explicit eyeball mesh for modeling corneal light‐transport, our model allows for animatable photorealistic 3D synthesis of the whole eye region. Using multi‐view video input, we demonstrate significant improvements over state‐of‐the‐art in expression re‐enactment and transfer for high‐resolution close‐up views of the eye region.

Despite recent advances in multi-view hair reconstruction, achieving strand-level precision remains a significant challenge due to inherent limitations in existing capture pipelines. We introduce GroomCap , a novel multi-view hair capture method that reconstructs faithful and high-fidelity hair geometry without relying on external data priors. To address the limitations of conventional reconstruction algorithms, we propose a neural implicit representation for hair volume that encodes high-resolution 3D orientation and occupancy from input views. This implicit hair volume is trained with a new volumetric 3D orientation rendering algorithm, coupled with 2D orientation distribution supervision, to effectively prevent the loss of structural information caused by undesired orientation blending. We further propose a Gaussian-based hair optimization strategy to refine the traced hair strands with a novel chained Gaussian representation, utilizing direct photometric supervision from images. Our results demonstrate that GroomCap is able to capture high-quality hair geometries that are not only more precise and detailed than existing methods but also versatile enough for a range of applications.

We propose a method to learn a high-quality implicit 3D head avatar from a monocular RGB video captured in the wild. The learnt avatar is driven by a parametric face model to achieve user-controlled facial expressions and head poses. Our hybrid pipeline combines the geometry prior and dynamic tracking of a 3DMM with a neural radiance field to achieve fine-grained control and photorealism. To reduce over-smoothing and improve out-of-model expressions synthesis, we propose to predict local features anchored on the 3DMM geometry. These learnt features are driven by 3DMM deformation and interpolated in 3D space to yield the volumetric radiance at a designated query point. We further show that using a Convolutional Neural Network in the UV space is critical in incorporating spatial context and producing representative local features. Extensive experiments show that we are able to reconstruct high-quality avatars, with more accurate expression-dependent details, good generalization to out-of-training expressions, and quantitatively superior renderings compared to other state-of-the-art approaches.

We present a novel method for populating 3D indoor scenes with virtual humans that can navigate in the environment and interact with objects in a realistic manner. Existing approaches rely on high-quality training sequences that contain captured human motions and the 3D scenes they interact with. However, such interaction data are costly, difficult to capture, and can hardly cover the full range of plausible human-scene interactions in complex indoor environments. To address these challenges, we propose a reinforcement learning-based approach that enables virtual humans to navigate in 3D scenes and interact with objects realistically and autonomously, driven by learned motion control policies. The motion control policies employ latent motion action spaces, which correspond to realistic motion primitives and are learned from large-scale motion capture data using a powerful generative motion model. For navigation in a 3D environment, we propose a scene-aware policy with novel state and reward designs for collision avoidance. Combined with navigation mesh-based path-finding algorithms to generate intermediate waypoints, our approach enables the synthesis of diverse human motions navigating in 3D indoor scenes and avoiding obstacles. To generate fine-grained human-object interactions, we carefully curate interaction goal guidance using a marker-based body representation and leverage features based on the signed distance field (SDF) to encode human-scene proximity relations. Our method can synthesize realistic and diverse human-object interactions (e.g., sitting on a chair and then getting up) even for out-of-distribution test scenarios with different object shapes, orientations, starting body positions, and poses. Experimental results demonstrate that our approach outperforms state-of-the-art human-scene interaction synthesis methods in terms of both motion naturalness and diversity. Code, models, and demonstrative video results are available at: https://zkf1997.github.io/DIMOS.

NeRFs have enabled highly realistic synthesis of human faces including complex appearance and reflectance effects of hair and skin. These methods typically require a large number of multi-view input images, making the process hardware intensive and cumbersome, limiting applicability to unconstrained settings. We propose a novel volumetric human face prior that enables the synthesis of ultra high-resolution novel views of subjects that are not part of the prior’s training distribution. This prior model consists of an identity-conditioned NeRF, trained on a dataset of low-resolution multi-view images of diverse humans with known camera calibration. A simple sparse landmark-based 3D alignment of the training dataset allows our model to learn a smooth latent space of geometry and appearance despite a limited number of training identities. A high-quality volumetric representation of a novel subject can be obtained by model fitting to 2 or 3 camera views of arbitrary resolution. Importantly, our method requires as few as two views of casually captured images as input at inference time.

We propose a novel transformer-based framework that reconstructs two high fidelity hands from multi-view RGB images. Unlike existing hand pose estimation methods, where one typically trains a deep network to regress hand model parameters from single RGB image, we consider a more challenging problem setting where we directly regress the absolute root poses of two-hands with extended forearm at high resolution from egocentric view. As existing datasets are either infeasible for egocentric viewpoints or lack background variations, we create a large-scale synthetic dataset with diverse scenarios and collect a real dataset from multi-calibrated camera setup to verify our proposed multi-view image feature fusion strategy. To make the reconstruction physically plausible, we propose two strategies: (i) a coarse-to-fine spectral graph convolution decoder to smoothen the meshes during upsampling and (ii) an optimisation-based refinement stage at inference to prevent self-penetrations. Through extensive quantitative and qualitative evaluations, we show that our framework is able to produce realistic two-hand reconstructions and demonstrate the generalisation of synthetic-trained models to real data, as well as real-time AR/VR applications.

Text-to-image generative models often reflect the biases of the training data, leading to unequal representations of underrepresented groups. This study investigates inclusive text-to-image generative models that generate images based on human-written prompts and ensure the resulting images are uniformly distributed across attributes of interest. Unfortunately, directly expressing the desired attributes in the prompt often leads to sub-optimal results due to linguistic ambiguity or model misrepresentation. Hence, this paper proposes a drastically different approach that adheres to the maxim that "a picture is worth a thousand words". We show that, for some attributes, images can represent concepts more expressively than text. For instance, categories of skin tones are typically hard to specify by text but can be easily represented by example images. Building upon these insights, we propose a novel approach, ITI-Gen1, that leverages readily available reference images for Inclusive Text-to-Image GENeration. The key idea is learning a set of prompt embeddings to generate images that can effectively represent all desired attribute categories. More importantly, ITI-Gen requires no model fine-tuning, making it computationally efficient to augment existing text-to-image models. Extensive experiments demonstrate that ITI-Gen largely improves over state-of-the-art models to generate inclusive images from a prompt.

High-fidelity, photorealistic 3D capture of a human face is a long-standing problem in computer graphics – the complex material of skin, intricate geometry of hair, and fine scale textural details make it challenging. Traditional techniques rely on very large and expensive capture rigs to reconstruct explicit mesh geometry and appearance maps, and are limited by the accuracy of hand-crafted reflectance models. More recent volumetric methods (e.g., NeRFs) have enabled view-synthesis and sometimes relighting by learning an implicit representation of the density and reflectance basis, but suffer from artifacts and blurriness due to the inherent ambiguities in volumetric modeling. These problems are further exacerbated when capturing with few cameras and light sources. We present a novel technique for high-quality capture of a human face for 3D view synthesis and relighting using a sparse, compact capture rig consisting of 15 cameras and 15 lights. Our method combines a neural volumetric representation with traditional mesh reconstruction from multiview stereo. The proxy geometry allows us to anchor the 3D density field to prevent artifacts and guide the disentanglement of intrinsic radiance components of the face appearance such as diffuse and specular reflectance, and incident radiance (shadowing) fields. Our hybrid representation significantly improves the state-of-the-art quality for arbitrarily dense renders of a face from desired camera viewpoint as well as environmental, directional, and near-field lighting.

Despite recent successes in hair acquisition that fits a high-dimensional hair model to a specific input subject, generative hair models, which establish general embedding spaces for encoding, editing, and sampling diverse hairstyles, are way less explored. In this paper, we present GroomGen , the first generative model designed for hair geometry composed of highly-detailed dense strands. Our approach is motivated by two key ideas. First, we construct hair latent spaces covering both individual strands and hairstyles. The latent spaces are compact, expressive, and well-constrained for high-quality and diverse sampling. Second, we adopt a hierarchical hair representation that parameterizes a complete hair model to three levels: single strands, sparse guide hairs, and complete dense hairs. This representation is critical to the compactness of latent spaces, the robustness of training, and the efficiency of inference. Based on this hierarchical latent representation, our proposed pipeline consists of a strand-VAE and a hairstyle-VAE that encode an individual strand and a set of guide hairs to their respective latent spaces, and a hybrid densification step that populates sparse guide hairs to a dense hair model. GroomGen not only enables novel hairstyle sampling and plausible hairstyle interpolation, but also supports interactive editing of complex hairstyles, or can serve as strong data-driven prior for hairstyle reconstruction from images. We demonstrate the superiority of our approach with qualitative examples of diverse sampled hairstyles and quantitative evaluation of generation quality regarding every single component and the entire pipeline.

A unique challenge in creating high-quality animatable and relightable 3D avatars of real people is modeling human eyes, particularly in conjunction with the surrounding periocular face region. The challenge of synthesizing eyes is multifold as it requires 1) appropriate representations for the various components of the eye and the periocular region for coherent viewpoint synthesis, capable of representing diffuse, refractive and highly reflective surfaces, 2) disentangling skin and eye appearance from environmental illumination such that it may be rendered under novel lighting conditions, and 3) capturing eyeball motion and the deformation of the surrounding skin to enable re-gazing. These challenges have traditionally necessitated the use of expensive and cumbersome capture setups to obtain high-quality results, and even then, modeling of the full eye region holistically has remained elusive. We present a novel geometry and appearance representation that enables high-fidelity capture and photorealistic animation, view synthesis and relighting of the eye region using only a sparse set of lights and cameras. Our hybrid representation combines an explicit parametric surface model for the eyeball surface with implicit deformable volumetric representations for the periocular region and the interior of the eye. This novel hybrid model has been designed specifically to address the various parts of that exceptionally challenging facial area - the explicit eyeball surface allows modeling refraction and high frequency specular reflection at the cornea, whereas the implicit representation is well suited to model lower frequency skin reflection via spherical harmonics and can represent non-surface structures such as hair (i.e. eyebrows) or highly diffuse volumetric bodies (i.e. sclera), both of which are a challenge for explicit surface models. Tightly integrating the two representations in a joint framework allows controlled photoreal image synthesis and joint optimization of both the geometry parameters of the eyeball and the implicit neural network in continuous 3D space. We show that for high-resolution close-ups of the human eye, our model can synthesize high-fidelity animated gaze from novel views under unseen illumination conditions, allowing to generate visually rich eye imagery.

Recent monocular human performance capture approaches have shown compelling dense tracking results of the full body from a single RGB camera. However, existing methods either do not estimate clothing at all or model cloth deformation with simple geometric priors instead of taking into account the underlying physical principles. This leads to noticeable artifacts in their reconstructions, e.g. baked-in wrinkles, implausible deformations that seemingly defy gravity, and intersections between cloth and body. To address these problems, we propose a person-specific, learning-based method that integrates a simulation layer into the training process to provide for the first time physics supervision in the context of weakly supervised deep monocular human performance capture. We show how integrating physics into the training process improves the learned cloth deformations, allows modeling clothing as a separate piece of geometry, and largely reduces cloth-body intersections. Relying only on weak 2D multi-view supervision during training, our approach leads to a significant improvement over current state-of-the-art methods and is thus a clear step towards realistic monocular capture of the entire deforming surface of a clothed human.

Deep generative models can synthesize photorealistic images of human faces with novel identities. However, a key challenge to the wide applicability of such techniques is to provide independent control over semantically meaningful parameters: appearance, head pose, face shape, and facial expressions. In this paper, we propose VariTex - to the best of our knowledge the first method that learns a variational latent feature space of neural face textures, which allows sampling of novel identities. We combine this generative model with a parametric face model and gain explicit control over head pose and facial expressions. To generate complete images of human heads, we propose an additive decoder that adds plausible details such as hair. A novel training scheme enforces a pose-independent latent space and in consequence, allows learning a one-to-many mapping between latent codes and pose-conditioned exterior regions. The resulting method can generate geometrically consistent images of novel identities under fine-grained control over head pose, face shape, and facial expressions. This facilitates a broad range of downstream tasks, like sampling novel identities, changing the head pose, expression transfer, and more.

Face models built from 3D face databases are often used in computer vision and graphics tasks such as face reconstruction, replacement, tracking and manipulation. For such tasks, commonly used multi-linear morphable models, which provide semantic control over facial identity and expression, often lack quality and expressivity due to their linear nature. Deep neural networks offer the possibility of non-linear face modeling, where so far most research has focused on generating realistic facial images with less focus on 3D geometry, and methods that do produce geometry have little or no notion of semantic control, thereby limiting their artistic applicability. We present a method for nonlinear 3D face modeling using neural architectures that provides intuitive semantic control over both identity and expression by disentangling these dimensions from each other, essentially combining the benefits of both multi-linear face models and nonlinear deep face networks. The result is a powerful, semantically controllable, nonlinear, parametric face model. We demonstrate the value of our semantic deep face model with applications of 3D face synthesis, facial performance transfer, performance editing, and 2D landmark-based performance retargeting.

Facial landmark detection is a fundamental task for many consumer and high-end applications and is almost entirely solved by machine learning methods today. Existing datasets used to train such algorithms are primarily made up of only low resolution images, and current algorithms are limited to inputs of comparable quality and resolution as the training dataset. On the other hand, high resolution imagery is becoming increasingly more common as consumer cameras improve in quality every year. Therefore, there is need for algorithms that can leverage the rich information available in high resolution imagery. Naively attempting to reuse existing network architectures on high resolution imagery is prohibitive due to memory bottlenecks on GPUs. The only current solution is to downsample the images, sacrificing resolution and quality. Building on top of recent progress in attention-based networks, we present a novel, fully convolutional regional architecture that is specially designed for predicting landmarks on very high resolution facial images without downsampling. We demonstrate the flexibility of our architecture by training the proposed model with images of resolutions ranging from 256 x 256 to 4K. In addition to being the first method for facial landmark detection on high resolution images, our approach achieves superior performance over traditional (holistic) state-of-the-art architectures across ALL resolutions, leading to a general-purpose, extremely flexible, high quality landmark detector.

Abstract Many problems in computer graphics and vision can be formulated as a nonlinear least squares optimization problem, for which numerous off‐the‐shelf solvers are readily available. Depending on the structure of the problem, however, existing solvers may be more or less suitable, and in some cases the solution comes at the cost of lengthy convergence times. One such case is semi‐sparse optimization problems, emerging for example in localized facial performance reconstruction, where the nonlinear least squares problem can be composed of hundreds of thousands of cost functions, each one involving many of the optimization parameters. While such problems can be solved with existing solvers, the computation time can severely hinder the applicability of these methods. We introduce a novel iterative solver for nonlinear least squares optimization of large‐scale semi‐sparse problems. We use the nonlinear Levenberg‐Marquardt method to locally linearize the problem in parallel, based on its first‐order approximation. Then, we decompose the linear problem in small blocks, using the local Schur complement, leading to a more compact linear system without loss of information. The resulting system is dense but its size is small enough to be solved using a parallel direct method in a short amount of time. The main benefit we get by using such an approach is that the overall optimization process is entirely parallel and scalable, making it suitable to be mapped onto graphics hardware (GPU). By using our minimizer, results are obtained up to one order of magnitude faster than other existing solvers, without sacrificing the generality and the accuracy of the model. We provide a detailed analysis of our approach and validate our results with the application of performance‐based facial capture using a recently‐proposed anatomical local face deformation model.

Abstract Digitally sculpting 3D human faces is a very challenging task. It typically requires either 1) highly‐skilled artists using complex software packages for high quality results, or 2) highly‐constrained simple interfaces for consumer‐level avatar creation, such as in game engines. We propose a novel interactive method for the creation of digital faces that is simple and intuitive to use, even for novice users, while consistently producing plausible 3D face geometry, and allowing editing freedom beyond traditional video game avatar creation. At the core of our system lies a specialized anatomical local face model (ALM), which is constructed from a dataset of several hundred 3D face scans. User edits are propagated to constraints for an optimization of our data‐driven ALM model, ensuring the resulting face remains plausible even for simple edits like clicking and dragging surface points. We show how several natural interaction methods can be implemented in our framework, including direct control of the surface, indirect control of semantic features like age, ethnicity, gender, and BMI, as well as indirect control through manipulating the underlying bony structures. The result is a simple new method for creating digital human faces, for artists and novice users alike. Our method is attractive for low‐budget VFX and animation productions, and our anatomical modeling paradigm can complement traditional game engine avatar design packages.

Abstract We present a practical method to synthesize plausible 3D facial expressions for a particular target subject. The ability to synthesize an entire facial rig from a single neutral expression has a large range of applications both in computer graphics and computer vision, ranging from the efficient and cost‐effective creation of CG characters to scalable data generation for machine learning purposes. Unlike previous methods based on multilinear models, the proposed approach is capable to extrapolate well outside the sample pool, which allows it to plausibly predict the identity of the target subject and create artifact free expression shapes while requiring only a small input dataset. We introduce global‐local multilinear models that leverage the strengths of expression‐specific and identity‐specific local models combined with coarse motion estimations from a global model. Experimental results show that we achieve high‐quality, plausible facial expression synthesis results for an individual that outperform existing methods both quantitatively and qualitatively.

Abstract 3D gaze tracking from a single RGB camera is very challenging due to the lack of information in determining the accurate gaze target from a monocular RGB sequence. The eyes tend to occupy only a small portion of the video, and even small errors in estimated eye orientations can lead to very large errors in the triangulated gaze target. We overcome these difficulties with a novel lightweight eyeball calibration scheme that determines the user‐specific visual axis, eyeball size and position in the head. Unlike the previous calibration techniques, we do not need the ground truth positions of the gaze points. In the online stage, gaze is tracked by a new gaze fitting algorithm, and refined by a 3D gaze regression method to correct for bias errors. Our regression is pre‐trained on several individuals and works well for novel users. After the lightweight one‐time user calibration, our method operates in real time. Experiments show that our technique achieves state‐of‐the‐art accuracy in gaze angle estimation, and we demonstrate applications of 3D gaze target tracking and gaze retargeting to an animated 3D character.

In this article, we provide a detailed survey of 3D Morphable Face Models over the 20 years since they were first proposed. The challenges in building and applying these models, namely, capture, modeling, image formation, and image analysis, are still active research topics, and we review the state-of-the-art in each of these areas. We also look ahead, identifying unsolved challenges, proposing directions for future research, and highlighting the broad range of current and future applications.

The increasing demand for 3D content in augmented and virtual reality has motivated the development of volumetric performance capture systemsnsuch as the Light Stage. Recent advances are pushing free viewpoint relightable videos of dynamic human performances closer to photorealistic quality. However, despite significant efforts, these sophisticated systems are limited by reconstruction and rendering algorithms which do not fully model complex 3D structures and higher order light transport effects such as global illumination and sub-surface scattering. In this paper, we propose a system that combines traditional geometric pipelines with a neural rendering scheme to generate photorealistic renderings of dynamic performances under desired viewpoint and lighting. Our system leverages deep neural networks that model the classical rendering process to learn implicit features that represent the view-dependent appearance of the subject independent of the geometry layout, allowing for generalization to unseen subject poses and even novel subject identity. Detailed experiments and comparisons demonstrate the efficacy and versatility of our method to generate high-quality results, significantly outperforming the existing state-of-the-art solutions.

We propose a new light-weight face capture system capable of reconstructing both high-quality geometry and detailed appearance maps from a single exposure. Unlike currently employed appearance acquisition systems, the proposed technology does not require active illumination and hence can readily be integrated with passive photogrammetry solutions. These solutions are in widespread use for 3D scanning humans as they can be assembled from off-the-shelf hardware components, but lack the capability of estimating appearance. This paper proposes a solution to overcome this limitation, by adding appearance capture to photogrammetry systems. The only additional hardware requirement to these solutions is that a subset of the cameras are cross-polarized with respect to the illumination, and the remaining cameras are parallel-polarized. The proposed algorithm leverages the images with the two different polarization states to reconstruct the geometry and to recover appearance properties. We do so by means of an inverse rendering framework, which solves per texel diffuse albedo, specular intensity, and high-resolution normals, as well as global specular roughness considering the subsurface scattering nature of skin. We show results for a variety of human subjects of different ages and skin typology, illustrating how the captured fine-detail skin surface and subsurface scattering effects lead to realistic renderings of their digital doubles, also in different illumination conditions.

Performance capture of expressive subjects, particularly facial performances acquired with high spatial resolution, will inevitably incorporate some fraction of motion that is due to inertial effects and dynamic overshoot due to ballistic motion. This is true in most natural capture environments where the actor is able to move freely during their performance, rather than being tethered to a fixed position. Normally these secondary dynamic effects are unwanted, as the captured facial performance is often retargeted to different head motion, and sometimes to completely different characters, and in both cases the captured dynamic effects should be removed and new secondary effects should be added. This paper advances the hypothesis that for a highly constrained elastic medium such as the human face, these secondary inertial effects are predominantly due to the motion of the underlying bony structures (cranium and mandible). Our work aims to compute and characterize the difference between the captured dynamic facial performance, and a speculative quasistatic variant of the same motion should the inertial effects have been absent. This is used to either subtract parasitic secondary dynamics that resulted from unintentional motion during capture, or compose such effects on top of a quasistatic performance to simulate a new dynamic motion of the actor's body and skull, either artist-prescribed or acquired via motion capture. We propose a data-driven technique that comprises complementary removal and synthesis networks for secondary dynamics in facial performance capture. We show how such a system can be effectively trained from a collection of acquired dynamic deformations under varying expressions where the actor induces rigid head motion from walking and running, as well as forced oscillatory body motion in a controlled setting by external actuators.

Abstract We present a novel parametric eye rig for eye animation, including a new multi‐view imaging system that can reconstruct eye poses at submillimeter accuracy to which we fit our new rig. This allows us to accurately estimate person‐specific eyeball shape, rotation center, interocular distance, visual axis, and other rig parameters resulting in an animation‐ready eye rig. We demonstrate the importance of several aspects of eye modeling that are often overlooked, for example that the visual axis is not identical to the optical axis, that it is important to model rotation about the optical axis, and that the rotation center of the eye should be measured accurately for each person. Since accurate rig fitting requires hand annotation of multi‐view imagery for several eye gazes, we additionally propose a more user‐friendly “lightweight” fitting approach, which leverages an average rig created from several pre‐captured accurate rigs. Our lightweight rig fitting method allows for the estimation of eyeball shape and eyeball position given only a single pose with a known look‐at point (e.g. looking into a camera) and few manual annotations.

Dubbing is a technique for translating video content from one language to another. However, state-of-the-art visual dubbing techniques directly copy facial expressions from source to target actors without considering identity-specific idiosyncrasies such as a unique type of smile. We present a style-preserving visual dubbing approach from single video inputs, which maintains the signature style of target actors when modifying facial expressions, including mouth motions, to match foreign languages. At the heart of our approach is the concept of motion style, in particular for facial expressions, i.e., the person-specific expression change that is yet another essential factor beyond visual accuracy in face editing applications. Our method is based on a recurrent generative adversarial network that captures the spatiotemporal co-activation of facial expressions, and enables generating and modifying the facial expressions of the target actor while preserving their style. We train our model with unsynchronized source and target videos in an unsupervised manner using cycle-consistency and mouth expression losses, and synthesize photorealistic video frames using a layered neural face renderer. Our approach generates temporally coherent results, and handles dynamic backgrounds. Our results show that our dubbing approach maintains the idiosyncratic style of the target actor better than previous approaches, even for widely differing source and target actors.

We present the first method to accurately track the invisible jaw based solely on the visible skin surface, without the need for any markers or augmentation of the actor. As such, the method can readily be integrated with off-the-shelf facial performance capture systems. The core idea is to learn a non-linear mapping from the skin deformation to the underlying jaw motion on a dataset where ground-truth jaw poses have been acquired, and then to retarget the mapping to new subjects. Solving for the jaw pose plays a central role in visual effects pipelines, since accurate jaw motion is required when retargeting to fantasy characters and for physical simulation. Currently, this task is performed mostly manually to achieve the desired level of accuracy, and the presented method has the potential to fully automate this labour intense and error prone process.

Abstract Facial performance capture is the primary method for generating facial animation in video games, feature films and virtual environments, and recent advances have produced very compelling results. Still, one of the most challenging regions is the mouth, which often contains systematic errors due to the complex appearance and occlusion/dis‐occlusion of the lips. We present a novel user‐guided approach to correcting these common lip shape errors present in traditional capture systems. Our approach is to allow a user to manually correct a small number of problematic frames, and then our system learns the types of corrections desired and automatically corrects the entire performance. As correcting even a single frame using traditional 3D sculpting tools can be time consuming and require great skill, we also propose a simple and fast 2D sketch‐based method for generating plausible lip corrections for the problematic key frames. We demonstrate our results on captured performances of three different subjects, and validate our method with an additional sequence that contains ground truth lip reconstructions.

Abstract The computer graphics and vision communities have dedicated long standing efforts in building computerized tools for reconstructing, tracking, and analyzing human faces based on visual input. Over the past years rapid progress has been made, which led to novel and powerful algorithms that obtain impressive results even in the very challenging case of reconstruction from a single RGB or RGB‐D camera. The range of applications is vast and steadily growing as these technologies are further improving in speed, accuracy, and ease of use. Motivated by this rapid progress, this state‐of‐the‐art report summarizes recent trends in monocular facial performance capture and discusses its applications, which range from performance‐based animation to real‐time facial reenactment. We focus our discussion on methods where the central task is to recover and track a three dimensional model of the human face using optimization‐based reconstruction algorithms. We provide an in‐depth overview of the underlying concepts of real‐world image formation, and we discuss common assumptions and simplifications that make these algorithms practical. In addition, we extensively cover the priors that are used to better constrain the under‐constrained monocular reconstruction problem, and discuss the optimization techniques that are employed to recover dense, photo‐geometric 3D face models from monocular 2D data. Finally, we discuss a variety of use cases for the reviewed algorithms in the context of motion capture, facial animation, as well as image and video editing.

We present a method to acquire dynamic properties of facial skin appearance, including dynamic diffuse albedo encoding blood flow, dynamic specular intensity, and per-frame high resolution normal maps for a facial performance sequence. The method reconstructs these maps from a purely passive multi-camera setup, without the need for polarization or requiring temporally multiplexed illumination. Hence, it is very well suited for integration with existing passive systems for facial performance capture. To solve this seemingly underconstrained problem, we demonstrate that albedo dynamics during a facial performance can be modeled as a combination of: (1) a static, high-resolution base albedo map, modeling full skin pigmentation; and (2) a dynamic, one-dimensional component in the CIE L*a*b* color space, which explains changes in hemoglobin concentration due to blood flow. We leverage this albedo subspace and additional constraints on appearance and surface geometry to also estimate specular reflection parameters and resolve high-resolution normal maps with unprecedented detail in a passive capture system. These constraints are built into an inverse rendering framework that minimizes the difference of the rendered face to the captured images, incorporating constraints from multiple views for every texel on the face. The presented method is the first system capable of capturing high-quality dynamic appearance maps at full resolution and video framerates, providing a major step forward in the area of facial appearance acquisition.

Recreating the appearance of humans in virtual environments for the purpose of movie, video game, or other types of production involves the acquisition of a geometric representation of the human body and its scattering parameters which express the interaction between the geometry and light propagated throughout the scene. Teeth appearance is defined not only by the light and surface interaction, but also by its internal geometry and the intra-oral environment, posing its own unique set of challenges. Therefore, we present a system specifically designed for capturing the optical properties of live human teeth such that they can be realistically re-rendered in computer graphics. We acquire our data in vivo in a conventional multiple camera and light source setup and use exact geometry segmented from intra-oral scans. To simulate the complex interaction of light in the oral cavity during inverse rendering we employ a novel pipeline based on derivative path tracing with respect to both optical properties and geometry of the inner dentin surface. The resulting estimates of the global derivatives are used to extract parameters in a joint numerical optimization. The final appearance faithfully recreates the acquired data and can be directly used in conventional path tracing frameworks for rendering virtual humans.

In computer graphics the motion of the jaw is commonly modelled by up-down and left-right rotation around a fixed pivot plus a forward-backward translation, yielding a three dimensional rig that is highly suited for intuitive artistic control. The anatomical motion of the jaw is, however, much more complex since the joints that connect the jaw to the skull exhibit both rotational and translational components. In reality the jaw does not move in a three dimensional subspace but on a constrained manifold in six dimensions. We analyze this manifold in the context of computer animation and show how the manifold can be parameterized with three degrees of freedom, providing a novel jaw rig that preserves the intuitive control while providing more accurate jaw positioning. The chosen parameterization furthermore places anatomically correct limits on the motion, preventing the rig from entering physiologically infeasible poses. Our new jaw rig is empirically designed from accurate capture data, and we provide a simple method to retarget the rig to new characters, both human and fantasy.

Calibrating the intrinsic properties of a camera is one of the fundamental tasks required for a variety of computer vision and image processing tasks. The precise measurement of focal length, location of the principal point as well as distortion parameters of the lens is crucial, for example, for 3D reconstruction [27]. Although a variety of methods exist to achieve this goal, they are often cumbersome to carry out, require substantial manual interaction, expert knowledge, and a significant operating volume. We propose a novel calibration method based on the usage of directionally encoded light rays for estimating the intrinsic parameters. It enables a fully automatic calibration with a small device mounted close to the front lens element and still enables an accuracy comparable to standard methods even when the lens is focused up to infinity. Our method overcomes the mentioned limitations since it guarantees an accurate calibration without any human intervention while requiring only a limited amount of space. Besides that, the approach also allows to estimate the distance of the focal plane as well as the size of the aperture. We demonstrate the advantages of the proposed method by evaluating several camera/lens configurations using prototypical devices.

Physical simulation has long been the approach of choice for generating realistic hair animations in CG. A constant drawback of simulation, however, is the necessity to manually set the physical parameters of the simulation model in order to get the desired dynamic behavior. To alleviate this, researchers have begun to explore methods for reconstructing hair from the real world and even to estimate the corresponding simulation parameters through the process of inversion. So far, however, these methods have had limited applicability, because dynamic hair capture can only be played back without the ability to edit, and solving for simulation parameters can only be accomplished for static hairstyles, ignoring the dynamic behavior. We present the first method for capturing dynamic hair and automatically determining the physical properties for simulating the observed hairstyle in motion. Since our dynamic inversion is agnostic to the simulation model, the proposed method applies to virtually any hair simulation technique, which we demonstrate using two state-of-the-art hair simulation models. The output of our method is a fully simulation-ready hairstyle, consisting of both the static hair geometry as well as its physical properties. The hairstyle can be easily edited by adding additional external forces, changing the head motion, or re-simulating in completely different environments, all while remaining faithful to the captured hairstyle.

We present a real-time multi-view facial capture system facilitated by synthetic training imagery. Our method is able to achieve high-quality markerless facial performance capture in real-time from multi-view helmet camera data, employing an actor specific regressor. The regressor training is tailored to specified actor appearance and we further condition it for the expected illumination conditions and the physical capture rig by generating the training data synthetically. In order to leverage the information present in live imagery, which is typically provided by multiple cameras, we propose a novel multi-view regression algorithm that uses multi-dimensional random ferns. We show that higher quality can be achieved by regressing on multiple video streams than previous approaches that were designed to operate on only a single view. Furthermore, we evaluate possible camera placements and propose a novel camera configuration that allows to mount cameras outside the field of view of the actor, which is very beneficial as the cameras are then less of a distraction for the actor and allow for an unobstructed line of sight to the director and other actors. Our new real-time facial capture approach has immediate application in on-set virtual production, in particular with the ever-growing demand for motion-captured facial animation in visual effects and video games.

Oftentimes facial animation is created separately from overall body motion. Since convincing facial animation is challenging enough in itself, artists tend to create and edit the face motion in isolation. Or if the face animation is derived from motion capture, this is typically performed in a mo-cap booth while sitting relatively still. In either case, recombining the isolated face animation with body and head motion is non-trivial and often results in an uncanny result if the body dynamics are not properly reflected on the face (e.g. the bouncing of facial tissue when running). We tackle this problem by introducing a simple and intuitive system that allows to add physics to facial blendshape animation. Unlike previous methods that try to add physics to face rigs, our method preserves the original facial animation as closely as possible. To this end, we present a novel simulation framework that uses the original animation as per-frame rest-poses without adding spurious forces. As a result, in the absence of any external forces or rigid head motion, the facial performance will exactly match the artist-created blendshape animation. In addition we propose the concept of blendmaterials to give artists an intuitive means to account for changing material properties due to muscle activation. This system allows to automatically combine facial animation and head motion such that they are consistent, while preserving the original animation as closely as possible. The system is easy to use and readily integrates with existing animation pipelines.

Real-time facial performance capture has recently been gaining popularity in virtual film production, driven by advances in machine learning, which allows for fast inference of facial geometry from video streams. These learning-based approaches are significantly influenced by the quality and amount of labelled training data. Tedious construction of training sets from real imagery can be replaced by rendering a facial animation rig under on-set conditions expected at runtime. We learn a synthetic actor-specific prior by adapting a state-of-the-art facial tracking method. Synthetic training significantly reduces the capture and annotation burden and in theory allows generation of an arbitrary amount of data. But practical realities such as training time and compute resources still limit the size of any training set. We construct better and smaller training sets by investigating which facial image appearances are crucial for tracking accuracy, covering the dimensions of expression, viewpoint and illumination. A reduction of training data in 1-2 orders of magnitude is demonstrated whilst tracking accuracy is retained for challenging on-set footage.

In facial animation, the accurate shape and motion of the lips of virtual humans is of paramount importance, since subtle nuances in mouth expression strongly influence the interpretation of speech and the conveyed emotion. Unfortunately, passive photometric reconstruction of expressive lip motions, such as a kiss or rolling lips, is fundamentally hard even with multi-view methods in controlled studios. To alleviate this problem, we present a novel approach for fully automatic reconstruction of detailed and expressive lip shapes along with the dense geometry of the entire face, from just monocular RGB video. To this end, we learn the difference between inaccurate lip shapes found by a state-of-the-art monocular facial performance capture approach, and the true 3D lip shapes reconstructed using a high-quality multi-view system in combination with applied lip tattoos that are easy to track. A robust gradient domain regressor is trained to infer accurate lip shapes from coarse monocular reconstructions, with the additional help of automatically extracted inner and outer 2D lip contours. We quantitatively and qualitatively show that our monocular approach reconstructs higher quality lip shapes, even for complex shapes like a kiss or lip rolling, than previous monocular approaches. Furthermore, we compare the performance of person-specific and multi-person generic regression strategies and show that our approach generalizes to new individuals and general scenes, enabling high-fidelity reconstruction even from commodity video footage.

Facial scanning has become ubiquitous in digital media, but so far most efforts have focused on reconstructing the skin. Eye reconstruction, on the other hand, has received only little attention, and the current state-of-the-art method is cumbersome for the actor, time-consuming, and requires carefully setup and calibrated hardware. These constraints currently make eye capture impractical for general use. We present the first approach for high-quality lightweight eye capture, which leverages a database of pre-captured eyes to guide the reconstruction of new eyes from much less constrained inputs, such as traditional single-shot face scanners or even a single photo from the internet. This is accomplished with a new parametric model of the eye built from the database, and a novel image-based model fitting algorithm. Our method provides both automatic reconstructions of real eyes, as well as artistic control over the parameters to generate user-specific eyes.

In recent years, sophisticated image-based reconstruction methods for the human face have been developed. These methods capture highly detailed static and dynamic geometry of the whole face, or specific models of face regions, such as hair, eyes or eye lids. Unfortunately, image-based methods to capture the mouth cavity in general, and the teeth in particular, have received very little attention. The accurate rendering of teeth, however, is crucial for the realistic display of facial expressions, and currently high quality face animations resort to tooth row models created by tedious manual work. In dentistry, special intra-oral scanners for teeth were developed, but they are invasive, expensive, cumbersome to use, and not readily available. In this paper, we therefore present the first approach for non-invasive reconstruction of an entire person-specific tooth row from just a sparse set of photographs of the mouth region. The basis of our approach is a new parametric tooth row prior learned from high quality dental scans. A new model-based reconstruction approach fits teeth to the photographs such that visible teeth are accurately matched and occluded teeth plausibly synthesized. Our approach seamlessly integrates into photogrammetric multi-camera reconstruction setups for entire faces, but also enables high quality teeth modeling from normal uncalibrated photographs and even short videos captured with a mobile phone.

We present a new anatomically-constrained local face model and fitting approach for tracking 3D faces from 2D motion data in very high quality. In contrast to traditional global face models, often built from a large set of blendshapes, we propose a local deformation model composed of many small subspaces spatially distributed over the face. Our local model offers far more flexibility and expressiveness than global blendshape models, even with a much smaller model size. This flexibility would typically come at the cost of reduced robustness, in particular during the under-constrained task of monocular reconstruction. However, a key contribution of this work is that we consider the face anatomy and introduce subspace skin thickness constraints into our model, which constrain the face to only valid expressions and helps counteract depth ambiguities in monocular tracking. Given our new model, we present a novel fitting optimization that allows 3D facial performance reconstruction from a single view at extremely high quality, far beyond previous fitting approaches. Our model is flexible, and can be applied also when only sparse motion data is available, for example with marker-based motion capture or even face posing from artistic sketches. Furthermore, by incorporating anatomical constraints we can automatically estimate the rigid motion of the skull, obtaining a rigid stabilization of the performance for free. We demonstrate our model and single-view fitting method on a number of examples, including, for the first time, extreme local skin deformation caused by external forces such as wind, captured from a single high-speed camera.

We present a method to continuously blend between multiple facial performances of an actor, which can contain different facial expressions or emotional states. As an example, given sad and angry video takes of a scene, our method empowers the movie director to specify arbitrary weighted combinations and smooth transitions between the two takes in post-production. Our contributions include (1) a robust nonlinear audio-visual synchronization technique that exploits complementary properties of audio and visual cues to automatically determine robust, dense spatiotemporal correspondences between takes, and (2) a seamless facial blending approach that provides the director full control to interpolate timing, facial expression, and local appearance, in order to generate novel performances after filming. In contrast to most previous works, our approach operates entirely in image space, avoiding the need of 3D facial reconstruction. We demonstrate that our method can synthesize visually believable performances with applications in emotion transition, performance correction, and timing control.

Abstract Facial appearance capture is now firmly established within academic research and used extensively across various application domains, perhaps most prominently in the entertainment industry through the design of virtual characters in video games and films. While significant progress has occurred over the last two decades, no single survey currently exists that discusses the similarities, differences, and practical considerations of the available appearance capture techniques as applied to human faces. A central difficulty of facial appearance capture is the way light interacts with skin—which has a complex multi‐layered structure—and the interactions that occur below the skin surface can, by definition, only be observed indirectly. In this report, we distinguish between two broad strategies for dealing with this complexity. “Image‐based methods” try to exhaustively capture the exact face appearance under different lighting and viewing conditions, and then render the face through weighted image combinations. “Parametric methods” instead fit the captured reflectance data to some parametric appearance model used during rendering, allowing for a more lightweight and flexible representation but at the cost of potentially increased rendering complexity or inexact reproduction. The goal of this report is to provide an overview that can guide practitioners and researchers in assessing the tradeoffs between current approaches and identifying directions for future advances in facial appearance capture.

In recent years we have seen numerous improvements on 3D scanning and tracking of human faces, greatly advancing the creation of digital doubles for film and video games. However, despite the high-resolution quality of the reconstruction approaches available, current methods are unable to capture one of the most important regions of the face - the eye region. In this work we present the first method for detailed spatio-temporal reconstruction of eyelids. Tracking and reconstructing eyelids is extremely challenging, as this region exhibits very complex and unique skin deformation where skin is folded under while opening the eye. Furthermore, eyelids are often only partially visible and obstructed due to self-occlusion and eyelashes. Our approach is to combine a geometric deformation model with image data, leveraging multi-view stereo, optical flow, contour tracking and wrinkle detection from local skin appearance. Our deformation model serves as a prior that enables reconstruction of eyelids even under strong self-occlusions caused by rolling and folding skin as the eye opens and closes. The output is a person-specific, time-varying eyelid reconstruction with anatomically plausible deformations. Our high-resolution detailed eyelids couple naturally with current facial performance capture approaches. As a result, our method can largely increase the fidelity of facial capture and the creation of digital doubles.

We present the first real-time high-fidelity facial capture method. The core idea is to enhance a global real-time face tracker, which provides a low-resolution face mesh, with local regressors that add in medium-scale details, such as expression wrinkles. Our main observation is that although wrinkles appear in different scales and at different locations on the face, they are locally very self-similar and their visual appearance is a direct consequence of their local shape. We therefore train local regressors from high-resolution capture data in order to predict the local geometry from local appearance at runtime. We propose an automatic way to detect and align the local patches required to train the regressors and run them efficiently in real-time. Our formulation is particularly designed to enhance the low-resolution global tracker with exactly the missing expression frequencies, avoiding superimposing spatial frequencies in the result. Our system is generic and can be applied to any real-time tracker that uses a global prior, e.g. blend-shapes. Once trained, our online capture approach can be applied to any new user without additional training, resulting in high-fidelity facial performance reconstruction with person-specific wrinkle details from a monocular video camera in real-time.

Facial scanning has become the industry-standard approach for creating digital doubles in movies and video games. This involves capturing an actor while they perform different expressions that span their range of facial motion. Unfortunately, the scans typically contain a superposition of the desired expression on top of un-wanted rigid head movement. In order to extract true expression deformations, it is essential to factor out the rigid head movement for each expression, a process referred to as rigid stabilization . In order to achieve production-quality in industry, face stabilization is usually performed through a tedious and error-prone manual process. In this paper we present the first automatic face stabilization method that achieves professional-quality results on large sets of facial expressions. Since human faces can undergo a wide range of deformation, there is not a single point on the skin surface that moves rigidly with the underlying skull. Consequently, computing the rigid transformation from direct observation, a common approach in previous methods, is error prone and leads to inaccurate results. Instead, we propose to indirectly stabilize the expressions by explicitly aligning them to an estimate of the underlying skull using anatomically-motivated constraints. We show that the proposed method not only outperforms existing techniques but is also on par with manual stabilization, yet requires less than a second of computation time.

Even though the human eye is one of the central features of individual appearance, its shape has so far been mostly approximated in our community with gross simplifications. In this paper we demonstrate that there is a lot of individuality to every eye, a fact that common practices for 3D eye generation do not consider. To faithfully reproduce all the intricacies of the human eye we propose a novel capture system that is capable of accurately reconstructing all the visible parts of the eye: the white sclera , the transparent cornea and the non-rigidly deforming colored iris . These components exhibit very different appearance properties and thus we propose a hybrid reconstruction method that addresses them individually, resulting in a complete model of both spatio-temporal shape and texture at an unprecedented level of detail, enabling the creation of more believable digital humans. Finally, we believe that the findings of this paper will alter our community's current assumptions regarding human eyes, and our work has the potential to significantly impact the way that eyes will be modelled in the future.

The facial performance of an individual is inherently rich in subtle deformation and timing details. Although these subtleties make the performance realistic and compelling, they often elude both motion capture and hand animation. We present a technique for adding fine-scale details and expressiveness to low-resolution art-directed facial performances, such as those created manually using a rig, via marker-based capture, by fitting a morphable model to a video, or through Kinect reconstruction using recent faceshift technology. We employ a high-resolution facial performance capture system to acquire a representative performance of an individual in which he or she explores the full range of facial expressiveness. From the captured data, our system extracts an expressiveness model that encodes subtle spatial and temporal deformation details specific to that particular individual. Once this model has been built, these details can be transferred to low-resolution art-directed performances. We demonstrate results on various forms of input; after our enhancement, the resulting animations exhibit the same nuances and fine spatial details as the captured performance, with optional temporal enhancement to match the dynamics of the actor. Finally, we show that our technique outperforms the current state-of-the-art in example-based facial animation.

Recently, we have seen a growing trend in the design and fabrication of personalized figurines, created by scanning real people and then physically reproducing miniature statues with 3D printers. This is currently a hot topic both in academia and industry, and the printed figurines are gaining more and more realism, especially with state-of-the-art facial scanning technology improving. However, current systems all contain the same limitation - no previous method is able to suitably capture personalized hair-styles for physical reproduction. Typically, the subject's hair is approximated very coarsely or replaced completely with a template model. In this paper we present the first method for stylized hair capture, a technique to reconstruct an individual's actual hair-style in a manner suitable for physical reproduction. Inspired by centuries-old artistic sculptures, our method generates hair as a closed-manifold surface, yet contains the structural and color elements stylized in a way that captures the defining characteristics of the hair-style. The key to our approach is a novel multi-view stylization algorithm, which extends feature-preserving color filtering from 2D images to irregular manifolds in 3D, and introduces abstract geometric details that are coherent with the color stylization. The proposed technique fits naturally in traditional pipelines for figurine reproduction, and we demonstrate the robustness and versatility of our approach by capturing several subjects with widely varying hair-styles.

Correspondence matching is one of the most common problems in computer vision, and it is often solved using photo-consistency of local regions. These approaches typically assume that the frequency content in the local region is consistent in the image pair, such that matching is performed on similar signals. However, in many practical situations this is not the case, for example with low depth of field cameras a scene point may be out of focus in one view and in-focus in the other, causing a mismatch of frequency signals. Furthermore, this mismatch can vary spatially over the entire image. In this paper we propose a local signal equalization approach for correspondence matching. Using a measure of local image frequency, we equalize local signals using an efficient scale-space image representation such that their frequency contents are optimally suited for matching. Our approach allows better correspondence matching, which we demonstrate with a number of stereo reconstruction examples on synthetic and real datasets.

Although facial hair plays an important role in individual expression, facial-hair reconstruction is not addressed by current face-capture systems. Our research addresses this limitation with an algorithm that treats hair and skin surface capture together in a coupled fashion so that a high-quality representation of hair fibers as well as the underlying skin surface can be reconstructed. We propose a passive, camera-based system that is robust against arbitrary motion since all data is acquired within the time period of a single exposure. Our reconstruction algorithm detects and traces hairs in the captured images and reconstructs them in 3D using a multiview stereo approach. Our coupled skin-reconstruction algorithm uses information about the detected hairs to deliver a skin surface that lies underneath all hairs irrespective of occlusions. In dense regions like eyebrows, we employ a hair-synthesis method to create hair fibers that plausibly match the image data. We demonstrate our scanning system on a number of individuals and show that it can successfully reconstruct a variety of facial-hair styles together with the underlying skin surface.

We propose a complete process for designing, simulating, and fabricating synthetic skin for an animatronics character that mimics the face of a given subject and its expressions. The process starts with measuring the elastic properties of a material used to manufacture synthetic soft tissue. Given these measurements we use physics-based simulation to predict the behavior of a face when it is driven by the underlying robotic actuation. Next, we capture 3D facial expressions for a given target subject. As the key component of our process, we present a novel optimization scheme that determines the shape of the synthetic skin as well as the actuation parameters that provide the best match to the target expressions. We demonstrate this computational skin design by physically cloning a real human face onto an animatronics figure.

We present a new technique for passive and markerless facial performance capture based on anchor frames . Our method starts with high resolution per-frame geometry acquisition using state-of-the-art stereo reconstruction, and proceeds to establish a single triangle mesh that is propagated through the entire performance. Leveraging the fact that facial performances often contain repetitive subsequences, we identify anchor frames as those which contain similar facial expressions to a manually chosen reference expression. Anchor frames are automatically computed over one or even multiple performances. We introduce a robust image-space tracking method that computes pixel matches directly from the reference frame to all anchor frames, and thereby to the remaining frames in the sequence via sequential matching. This allows us to propagate one reconstructed frame to an entire sequence in parallel, in contrast to previous sequential methods. Our anchored reconstruction approach also limits tracker drift and robustly handles occlusions and motion blur. The parallel tracking and mesh propagation offer low computation times. Our technique will even automatically match anchor frames across different sequences captured on different occasions, propagating a single mesh to all performances.

This paper describes a passive stereo system for capturing the 3D geometry of a face in a single-shot under standard light sources. The system is low-cost and easy to deploy. Results are submillimeter accurate and commensurate with those from state-of-the-art systems based on active lighting, and the models meet the quality requirements of a demanding domain like the movie industry. Recovered models are shown for captures from both high-end cameras in a studio setting and from a consumer binocular-stereo camera, demonstrating scalability across a spectrum of camera deployments, and showing the potential for 3D face modeling to move beyond the professional arena and into the emerging consumer market in stereoscopic photography. Our primary technical contribution is a modification of standard stereo refinement methods to capture pore-scale geometry, using a qualitative approach that produces visually realistic results. The second technical contribution is a calibration method suited to face capture systems. The systemic contribution includes multiple demonstrations of system robustness and quality. These include capture in a studio setup, capture off a consumer binocular-stereo camera, scanning of faces of varying gender and ethnicity and age, capture of highly-transient facial expression, and scanning a physical mask to provide ground-truth validation.