This paper introduces Comprehensive Relighting, the first all-in-one approach that can both control and harmonize the lighting from an image or video of humans with arbitrary body parts from any scene. Building such a generalizable model is extremely challenging due to the lack of dataset, restricting existing image-based relighting models to a specific scenario (e.g., face or static human). To address this challenge, we repurpose a pre-trained diffusion model as a general image prior and jointly model the human relighting and background harmonization in the coarse-to-fine framework. To further enhance the temporal coherence of the relighting, we introduce an unsupervised temporal lighting model that learns the lighting cycle consistency from many real-world videos without any ground truth. In inference time, our temporal lighting module is combined with the diffusion models through the spatio-temporal feature blending algorithms without extra training; and we apply a new guided refinement as a post-processing to pre-serve the high-frequency details from the input image. In the experiments, Comprehensive Relighting shows a strong generalizability and lighting temporal coherence, outperforming existing image-based human relighting and harmonization methods.

Graphic design is crucial for conveying ideas and messages. Designers usually organize their work into objects, backgrounds, and vectorized text layers to simplify editing. However, this workflow demands considerable expertise. With the rise of GenAI methods, an endless supply of high-quality graphic designs in pixel format has become more accessible, though these designs often lack editability. Despite this, non-layered designs still inspire human designers, influencing their choices in layouts and text styles, ultimately guiding the creation of layered designs. Motivated by this observation, we propose Accordion, a graphic design generation framework taking the first attempt to convert AI-generated designs into editable layered designs, meanwhile refining nonsensical AI-generated text with meaningful alternatives guided by user prompts. It is built around a vision language model (VLM) playing distinct roles in three curated stages. For each stage, we design prompts to guide the VLM in executing different tasks. Distinct from existing bottom-up methods (e.g., COLE and Open-COLE) that gradually generate elements to create layered designs, our approach works in a top-down manner by using the visually harmonious reference image as global guidance to decompose each layer. Additionally, it leverages multiple vision experts such as SAM and element removal models to facilitate the creation of graphic layers. We train our method using the in-house graphic design dataset Design39K, augmented with AI-generated design images coupled with refined ground truth created by a customized inpainting model. Experimental results and user studies by designers show that Accordion generates favorable results on the DesignIntention benchmark, including tasks such as text-to-template, adding text to background, and text de-rendering, and also excels in creating design variations.

This paper proposes an image-vector dual diffusion model for generative layout design. Distinct from prior efforts that mostly ignores visual information of elements and the whole canvas, our approach integrates the power of a pre-trained large image diffusion model to guide layout composition in a vector diffusion model by providing enhanced salient region understanding and high-level inter-element relationship reasoning. Our proposed model simultaneously operates in two domains: it generates the overall design appearance in the image domain while optimizing the size and position of each design element in the vector domain. The proposed method achieves the state-of-the-art results on several datasets and enables new layout design applications. Project webpage: https://aminshabani.github.io/visual_layout_composer.

Remarkable strides have been made in reconstructing static scenes or human bodies from monocular videos. Yet, the two problems have largely been approached independently, without much synergy. Most visual SLAM methods can only reconstruct camera trajectories and scene structures up to scale, while most HMR methods reconstruct human meshes in metric scale but fall short in reasoning with cameras and scenes. This work introduces Synergistic Camera and Human Reconstruction (SynCHMR) to marry the best of both worlds. Specifically, we design Human-aware Metric SLAM to reconstruct metric-scale camera poses and scene point clouds using camera-frame HMR as a strong prior, addressing depth, scale, and dynamic ambiguities. Conditioning on the dense scene recovered, we further learn a Scene-aware SMPL Denoiser to enhance world-frame HMR by incorporating spatio-temporal coherency and dynamic scene constraints. Together, they lead to consistent reconstructions of camera trajectories, human meshes, and dense scene point clouds in a common world frame.

We study a novel problem to automatically generate video background that tailors to foreground subject motion. It is an important problem for the movie industry and visual effects community, which traditionally requires tedious manual efforts to solve. To this end, we propose ActAnywhere, a video diffusion model that takes as input a sequence of foreground subject segmentation and an image of a novel background and generates a video of the subject interacting in this background. We train our model on a large-scale dataset of 2.4M videos of human-scene interactions. Through extensive evaluation, we show that our model produces videos with realistic foreground-background interaction while strictly following the guidance of the condition image. Our model generalizes to diverse scenarios including non-human subjects, gaming and animation clips, as well as videos with multiple moving subjects. Both quantitative and qualitative comparisons demonstrate that our model significantly outperforms existing methods, which fail to accomplish the studied task. Please visit our project webpage at https://actanywhere.github.io.

Clothes undergo complex geometric deformations, which lead to appearance changes. To edit human videos in a physically plausible way, a texture map must take into account not only the garment transformation induced by the body movements and clothes fitting, but also its 3D fine-grained surface geometry. This poses, however, a new challenge of 3D reconstruction of dynamic clothes from an image or a video. In this paper, we show that it is possible to edit dressed human images and videos without 3D reconstruction. We estimate a geometry aware texture map between the garment region in an image and the texture space, a.k.a, UV map. Our UV map is designed to preserve isometry with respect to the underlying 3D surface by making use of the 3D surface normals predicted from the image. Our approach captures the underlying geometry of the garment in a self-supervised way, requiring no ground truth annotation of UV maps and can be readily extended to predict temporally coherent UV maps. We demonstrate that our method outperforms the state-of-the-art human UV map estimation approaches on both real and synthetic data.

We study the problem of inferring scene affordances by presenting a method for realistically inserting people into scenes. Given a scene image with a marked region and an image of a person, we insert the person into the scene while respecting the scene affordances. Our model can infer the set of realistic poses given the scene context, re-pose the reference person, and harmonize the composition. We set up the task in a self-supervised fashion by learning to repose humans in video clips. We train a large-scale diffusion model on a dataset of 2.4M video clips that produces diverse plausible poses while respecting the scene context. Given the learned human-scene composition, our model can also hallucinate realistic people and scenes when prompted without conditioning and also enables interactive editing. A quantitative evaluation shows that our method synthesizes more realistic human appearance and more natural human-scene interactions than prior work.

This paper presents a novel object-centric contact representation ContactGen for hand-object interaction. The ContactGen comprises 3 components: a contact map indicates the contact location, a part map represents the contact hand part, and a direction map tells the contact direction within each part. Given an input object, we propose a conditional generative model to predict ContactGen and adopt model-based optimization to predict diverse and geometrically feasible grasps. Experimental results demonstrate our method can generate high-fidelity and diverse human grasps for various objects.

Appearance of dressed humans undergoes a complex geometric transformation induced not only by the static pose but also by its dynamics, i.e., there exists a number of cloth geometric configurations given a pose depending on the way it has moved. Such appearance modeling conditioned on motion has been largely neglected in existing human rendering methods, resulting in rendering of physically implausible motion. A key challenge of learning the dynamics of the appearance lies in the requirement of a prohibitively large amount of observations. In this paper, we present a compact motion representation by enforcing equivariance—a representation is expected to be transformed in the way that the pose is transformed. We model an equivariant encoder that can generate the generalizable representation from the spatial and temporal derivatives of the 3D body surface. This learned representation is decoded by a compositional multi-task decoder that renders high fidelity time-varying appearance. Our experiments show that our method can generate a temporally coherent video of dynamic humans for unseen body poses and novel views given a single view video.

Human speech is often accompanied by body gestures including arm and hand gestures. We present a method that reenacts a high-quality video with gestures matching a target speech audio. The key idea of our method is to split and re-assemble clips from a reference video through a novel video motion graph encoding valid transitions between clips. To seamlessly connect different clips in the reenactment, we propose a pose-aware video blending network which synthesizes video frames around the stitched frames between two clips. Moreover, we developed an audio-based gesture searching algorithm to find the optimal order of the reenacted frames. Our system generates reen-actments that are consistent with both the audio rhythms and the speech content. We evaluate our synthesized video quality quantitatively, qualitatively, and with user studies, demonstrating that our method produces videos of much higher quality and consistency with the target audio compared to previous work and baselines. Our project page https://github.com/yzhou359/vid-reenact includes code and data.

A deep generative model that describes human motions can benefit a wide range of fundamental computer vision and graphics tasks, such as providing robustness to video-based human pose estimation, predicting complete body movements for motion capture systems during occlusions, and assisting key frame animation with plausible movements. In this paper, we present a method for learning complex human motions independent of specific tasks using a combined global and local latent space to facilitate coarse and fine-grained modeling. Specifically, we propose a hierarchical motion variational autoencoder (HM-VAE) that consists of a 2-level hierarchical latent space. While the global latent space captures the overall global body motion, the local latent space enables to capture the refined poses of the different body parts. We demonstrate the effectiveness of our hierarchical motion variational autoencoder in a variety of tasks including video-based human pose estimation, motion completion from partial observations, and motion synthesis from sparse key-frames. Even though, our model has not been trained for any of these tasks specifically, it provides superior performance than task-specific alternatives. Our general-purpose human motion prior model can fix corrupted human body animations and generate complete movements from incomplete observations.

Recent progress on physics-based character animation has shown impressive breakthroughs on human motion synthesis, through imitating motion capture data via deep reinforcement learning. However, results have mostly been demonstrated on imitating a single distinct motion pattern, and do not generalize to interactive tasks that require flexible motion patterns due to varying human-object spatial configurations. To bridge this gap, we focus on one class of interactive tasks---sitting onto a chair. We propose a hierarchical reinforcement learning framework which relies on a collection of subtask controllers trained to imitate simple, reusable mocap motions, and a meta controller trained to execute the subtasks properly to complete the main task. We experimentally demonstrate the strength of our approach over different non-hierarchical and hierarchical baselines. We also show that our approach can be applied to motion prediction given an image input. A supplementary video can be found at https://youtu.be/3CeN0OGz2cA.

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

We introduce HuMoR: a 3D Human Motion Model for Robust Estimation of temporal pose and shape. Though substantial progress has been made in estimating 3D human motion and shape from dynamic observations, recovering plausible pose sequences in the presence of noise and occlusions remains a challenge. For this purpose, we propose an expressive generative model in the form of a conditional variational autoencoder, which learns a distribution of the change in pose at each step of a motion sequence. Furthermore, we introduce a flexible optimization-based approach that leverages HuMoR as a motion prior to robustly estimate plausible pose and shape from ambiguous observations. Through extensive evaluations, we demonstrate that our model generalizes to diverse motions and body shapes after training on a large motion capture dataset, and enables motion reconstruction from multiple input modalities including 3D keypoints and RGB(-D) videos. See the project page at geometry.stanford.edu/projects/humor.

This paper introduces a motion retargeting method that preserves self-contacts and prevents interpenetration. Self-contacts, such as when hands touch each other or the torso or the head, are important attributes of human body language and dynamics, yet existing methods do not model or preserve these contacts. Likewise, interpenetration, such as a hand passing into the torso, are a typical artifact of motion estimation methods. The input to our method is a human motion sequence and a target skeleton and character geometry. The method identifies self-contacts and ground contacts in the input motion, and optimizes the motion to apply to the output skeleton, while preserving these contacts and reducing interpenetration. We introduce a novel geometry-conditioned recurrent network with an encoder-space optimization strategy that achieves efficient retargeting while satisfying contact constraints. In experiments, our results quantitatively outperform previous methods and we conduct a user study where our retargeted motions are rated as higher-quality than those produced by recent works. We also show our method generalizes to motion estimated from human videos where we improve over previous works that produce noticeable interpenetration.

Layout is important for graphic design and scene generation. We propose a novel Generative Adversarial Network, called LayoutGAN, that synthesizes layouts by modeling geometric relations of different types of 2D elements. The generator of LayoutGAN takes as input a set of randomly-placed 2D graphic elements, represented by vectors and uses self-attention modules to refine their labels and geometric parameters jointly to produce a realistic layout. Accurate alignment is critical for good layouts. We, thus, propose a novel differentiable wireframe rendering layer that maps the generated layout to a wireframe image, upon which a CNN-based discriminator is used to optimize the layouts in image space. We validate the effectiveness of LayoutGAN in various experiments including MNIST digit generation, document layout generation, clipart abstract scene generation, tangram graphic design, mobile app layout design, and webpage layout optimization from hand-drawn sketches.

We present an algorithm for re-rendering a person from a single image under arbitrary poses. Existing methods often have difficulties in hallucinating occluded contents photo-realistically while preserving the identity and fine details in the source image. We first learn to inpaint the correspondence field between the body surface texture and the source image with a human body symmetry prior. The inpainted correspondence field allows us to transfer/warp local features extracted from the source to the target view even under large pose changes. Directly mapping the warped local features to an RGB image using a simple CNN decoder often leads to visible artifacts. Thus, we extend the StyleGAN generator so that it takes pose as input (for controlling poses) and introduces a spatially varying modulation for the latent space using the warped local features (for controlling appearances). We show that our method compares favorably against the state-of-the-art algorithms in both quantitative evaluation and visual comparison.

Modeling layout is an important first step for graphic design. Recently, methods for generating graphic layouts have progressed, particularly with Generative Adversarial Networks (GANs). However, the problem of specifying the locations and sizes of design elements usually involves constraints with respect to element attributes, such as area, aspect ratio and reading-order. Automating attribute conditional graphic layouts remains a complex and unsolved problem. In this article, we introduce Attribute-conditioned Layout GAN to incorporate the attributes of design elements for graphic layout generation by forcing both the generator and the discriminator to meet attribute conditions. Due to the complexity of graphic designs, we further propose an element dropout method to make the discriminator look at partial lists of elements and learn their local patterns. In addition, we introduce various loss designs following different design principles for layout optimization. We demonstrate that the proposed method can synthesize graphic layouts conditioned on different element attributes. It can also adjust well-designed layouts to new sizes while retaining elements' original reading-orders. The effectiveness of our method is validated through a user study.

In this paper, we aim to reduce the footskate artifacts when reconstructing human dynamics from monocular RGB videos. Recent work has made substantial progress in improving the temporal smoothness of the reconstructed motion trajectories. Their results, however, still suffer from severe foot skating and slippage artifacts. To tackle this issue, we present a neural network based detector for localizing ground contact events of human feet and use it to impose a physical constraint for optimization of the whole human dynamics in a video. We present a detailed study on the proposed ground contact detector and demonstrate high-quality human motion reconstruction results in various videos.

Abstract Plausible conversations among characters are required to generate the ambiance of social settings such as a restaurant, hotel lobby, or cocktail party. In this paper, we propose a motion synthesis technique that can rapidly generate animated motion for characters engaged in two‐party conversations. Our system synthesizes gestures and other body motions for dyadic conversations that synchronize with novel input audio clips. Human conversations feature many different forms of coordination and synchronization. For example, speakers use hand gestures to emphasize important points, and listeners often nod in agreement or acknowledgment. To achieve the desired degree of realism, our method first constructs a motion graph that preserves the statistics of a database of recorded conversations performed by a pair of actors. This graph is then used to search for a motion sequence that respects three forms of audio‐motion coordination in human conversations: coordination to phonemic clause, listener response, and partner's hesitation pause. We assess the quality of the generated animations through a user study that compares them to the originally recorded motion and evaluate the effects of each type of audio‐motion coordination via ablation studies.

Splashing is one of the most fascinating liquid phenomena in the real world and it is favored by artists to create stunning visual effects, both statically and dynamically. Unfortunately, the generation of complex and specialized liquid splashes is a challenging task and often requires considerable time and effort. In this paper, we present a novel system that synthesizes realistic liquid splashes from simple user sketch input. Our system adopts a conditional generative adversarial network (cGAN) trained with physics-based simulation data to produce raw liquid splash models from input sketches, and then applies model refinement processes to further improve their small-scale details. The system considers not only the trajectory of every user stroke, but also its speed, which makes the splash model simulation-ready with its underlying 3D flow. Compared with simulation-based modeling techniques through trials and errors, our system offers flexibility, convenience and intuition in liquid splash design and editing. We evaluate the usability and the efficiency of our system in an immersive virtual reality environment. Thanks to this system, an amateur user can now generate a variety of realistic liquid splashes in just a few minutes.

Existing image inpainting methods typically fill holes by borrowing information from surrounding pixels. They often produce unsatisfactory results when the holes overlap with or touch foreground objects due to lack of information about the actual extent of foreground and background regions within the holes. These scenarios, however, are very important in practice, especially for applications such as distracting object removal. To address the problem, we propose a foreground-aware image inpainting system that explicitly disentangles structure inference and content completion. Specifically, our model learns to predict the foreground contour first, and then inpaints the missing region using the predicted contour as guidance. We show that by such disentanglement, the contour completion model predicts reasonable contours of objects, and further substantially improves the performance of image inpainting. Experiments show that our method significantly outperforms existing methods and achieves superior inpainting results on challenging cases with complex compositions.

In neural networks, it is often desirable to work with various representations of the same space. For example, 3D rotations can be represented with quaternions or Euler angles. In this paper, we advance a definition of a continuous representation, which can be helpful for training deep neural networks. We relate this to topological concepts such as homeomorphism and embedding. We then investigate what are continuous and discontinuous representations for 2D, 3D, and n-dimensional rotations. We demonstrate that for 3D rotations, all representations are discontinuous in the real Euclidean spaces of four or fewer dimensions. Thus, widely used representations such as quaternions and Euler angles are discontinuous and difficult for neural networks to learn. We show that the 3D rotations have continuous representations in 5D and 6D, which are more suitable for learning. We also present continuous representations for the general case of the n-dimensional rotation group SO(n). While our main focus is on rotations, we also show that our constructions apply to other groups such as the orthogonal group and similarity transforms. We finally present empirical results, which show that our continuous rotation representations outperform discontinuous ones for several practical problems in graphics and vision, including a simple autoencoder sanity test, a rotation estimator for 3D point clouds, and an inverse kinematics solver for 3D human poses.

We present a generative image inpainting system to complete images with free-form mask and guidance. The system is based on gated convolutions learned from millions of images without additional labelling efforts. The proposed gated convolution solves the issue of vanilla convolution that treats all input pixels as valid ones, generalizes partial convolution by providing a learnable dynamic feature selection mechanism for each channel at each spatial location across all layers. Moreover, as free-form masks may appear anywhere in images with any shape, global and local GANs designed for a single rectangular mask are not applicable. Thus, we also present a patch-based GAN loss, named SN-PatchGAN, by applying spectral-normalized discriminator on dense image patches. SN-PatchGAN is simple in formulation, fast and stable in training. Results on automatic image inpainting and user-guided extension demonstrate that our system generates higher-quality and more flexible results than previous methods. Our system helps user quickly remove distracting objects, modify image layouts, clear watermarks and edit faces. Code, demo and models are available at: \url{https://github.com/JiahuiYu/generative_inpainting}.

An assumption widely used in recent neural style transfer methods is that image styles can be described by global statics of deep features like Gram or covariance matrices. Alternative approaches have represented styles by decomposing them into local pixel or neural patches. Despite the recent progress, most existing methods treat the semantic patterns of style image uniformly, resulting unpleasing results on complex styles. In this paper, we introduce a more flexible and general universal style transfer technique: multimodal style transfer (MST). MST explicitly considers the matching of semantic patterns in content and style images. Specifically, the style image features are clustered into sub-style components, which are matched with local content features under a graph cut formulation. A reconstruction network is trained to transfer each sub-style and render the final stylized result. We also generalize MST to improve some existing methods. Extensive experiments demonstrate the superior effectiveness, robustness, and flexibility of MST.

Estimating 3D hand pose from single RGB images is a highly ambiguous problem that relies on an unbiased training dataset. In this paper, we analyze cross-dataset generalization when training on existing datasets. We find that approaches perform well on the datasets they are trained on, but do not generalize to other datasets or in-the-wild scenarios. As a consequence, we introduce the first large-scale, multi-view hand dataset that is accompanied by both 3D hand pose and shape annotations. For annotating this real-world dataset, we propose an iterative, semi-automated `human-in-the-loop' approach, which includes hand fitting optimization to infer both the 3D pose and shape for each sample. We show that methods trained on our dataset consistently perform well when tested on other datasets. Moreover, the dataset allows us to train a network that predicts the full articulated hand shape from a single RGB image. The evaluation set can serve as a benchmark for articulated hand shape estimation.

The Ken Burns effect allows animating still images with a virtual camera scan and zoom. Adding parallax, which results in the 3D Ken Burns effect, enables significantly more compelling results. Creating such effects manually is time-consuming and demands sophisticated editing skills. Existing automatic methods, however, require multiple input images from varying viewpoints. In this paper, we introduce a framework that synthesizes the 3D Ken Burns effect from a single image, supporting both a fully automatic mode and an interactive mode with the user controlling the camera. Our framework first leverages a depth prediction pipeline, which estimates scene depth that is suitable for view synthesis tasks. To address the limitations of existing depth estimation methods such as geometric distortions, semantic distortions, and inaccurate depth boundaries, we develop a semantic-aware neural network for depth prediction, couple its estimate with a segmentation-based depth adjustment process, and employ a refinement neural network that facilitates accurate depth predictions at object boundaries. According to this depth estimate, our framework then maps the input image to a point cloud and synthesizes the resulting video frames by rendering the point cloud from the corresponding camera positions. To address disocclusions while maintaining geometrically and temporally coherent synthesis results, we utilize context-aware color- and depth-inpainting to fill in the missing information in the extreme views of the camera path, thus extending the scene geometry of the point cloud. Experiments with a wide variety of image content show that our method enables realistic synthesis results. Our study demonstrates that our system allows users to achieve better results while requiring little effort compared to existing solutions for the 3D Ken Burns effect creation.

This paper proposes a deep neural network (DNN) for piece-wise planar depthmap reconstruction from a single RGB image. While DNNs have brought remarkable progress to single-image depth prediction, piece-wise planar depthmap reconstruction requires a structured geometry representation, and has been a difficult task to master even for DNNs. The proposed end-to-end DNN learns to directly infer a set of plane parameters and corresponding plane segmentation masks from a single RGB image. We have generated more than 50,000 piece-wise planar depthmaps for training and testing from ScanNet, a large-scale RGBD video database. Our qualitative and quantitative evaluations demonstrate that the proposed approach outperforms baseline methods in terms of both plane segmentation and depth estimation accuracy. To the best of our knowledge, this paper presents the first end-to-end neural architecture for piece-wise planar reconstruction from a single RGB image. Code and data are available at https://github.com/art-programmer/PlaneNet.

We propose a recurrent neural network architecture with a Forward Kinematics layer and cycle consistency based adversarial training objective for unsupervised motion retargetting. Our network captures the high-level properties of an input motion by the forward kinematics layer, and adapts them to a target character with different skeleton bone lengths (e.g., shorter, longer arms etc.). Collecting paired motion training sequences from different characters is expensive. Instead, our network utilizes cycle consistency to learn to solve the Inverse Kinematics problem in an unsupervised manner. Our method works online, i.e., it adapts the motion sequence on-the-fly as new frames are received. In our experiments, we use the Mixamo animation data1 to test our method for a variety of motions and characters and achieve state-of-the-art results. We also demonstrate motion retargetting from monocular human videos to 3D characters using an off-the-shelf 3D pose estimator.

In this paper, we address referring expression comprehension: localizing an image region described by a natural language expression. While most recent work treats expressions as a single unit, we propose to decompose them into three modular components related to subject appearance, location, and relationship to other objects. This allows us to flexibly adapt to expressions containing different types of information in an end-to-end framework. In our model, which we call the Modular Attention Network (MAttNet), two types of attention are utilized: language-based attention that learns the module weights as well as the word/phrase attention that each module should focus on; and visual attention that allows the subject and relationship modules to focus on relevant image components. Module weights combine scores from all three modules dynamically to output an overall score. Experiments show that MAttNet outperforms previous state-of-the-art methods by a large margin on both bounding-box-level and pixel-level comprehension tasks. Demo1 and code2 are provided.

Recent deep learning based approaches have shown promising results for the challenging task of inpainting large missing regions in an image. These methods can generate visually plausible image structures and textures, but often create distorted structures or blurry textures inconsistent with surrounding areas. This is mainly due to ineffectiveness of convolutional neural networks in explicitly borrowing or copying information from distant spatial locations. On the other hand, traditional texture and patch synthesis approaches are particularly suitable when it needs to borrow textures from the surrounding regions. Motivated by these observations, we propose a new deep generative model-based approach which can not only synthesize novel image structures but also explicitly utilize surrounding image features as references during network training to make better predictions. The model is a feedforward, fully convolutional neural network which can process images with multiple holes at arbitrary locations and with variable sizes during the test time. Experiments on multiple datasets including faces (CelebA, CelebA-HQ), textures (DTD) and natural images (ImageNet, Places2) demonstrate that our proposed approach generates higher-quality inpainting results than existing ones. Code, demo and models are available at: https://github.com/JiahuiYu/generative_inpainting.

This paper presents the first study on forecasting human dynamics from static images. The problem is to input a single RGB image and generate a sequence of upcoming human body poses in 3D. To address the problem, we propose the 3D Pose Forecasting Network (3D-PFNet). Our 3D-PFNet integrates recent advances on single-image human pose estimation and sequence prediction, and converts the 2D predictions into 3D space. We train our 3D-PFNet using a three-step training strategy to leverage a diverse source of training data, including image and video based human pose datasets and 3D motion capture (MoCap) data. We demonstrate competitive performance of our 3D-PFNet on 2D pose forecasting and 3D structure recovery through quantitative and qualitative results.

Recent progresses on deep discriminative and generative modeling have shown promising results on texture synthesis. However, existing feed-forward based methods trade off generality for efficiency, which suffer from many issues, such as shortage of generality (i.e., build one network per texture), lack of diversity (i.e., always produce visually identical output) and suboptimality (i.e., generate less satisfying visual effects). In this work, we focus on solving these issues for improved texture synthesis. We propose a deep generative feed-forward network which enables efficient synthesis of multiple textures within one single network and meaningful interpolation between them. Meanwhile, a suite of important techniques are introduced to achieve better convergence and diversity. With extensive experiments, we demonstrate the effectiveness of the proposed model and techniques for synthesizing a large number of textures and show its applications with the stylization.

In this paper, we propose an effective face completion algorithm using a deep generative model. Different from well-studied background completion, the face completion task is more challenging as it often requires to generate semantically new pixels for the missing key components (e.g., eyes and mouths) that contain large appearance variations. Unlike existing nonparametric algorithms that search for patches to synthesize, our algorithm directly generates contents for missing regions based on a neural network. The model is trained with a combination of a reconstruction loss, two adversarial losses and a semantic parsing loss, which ensures pixel faithfulness and local-global contents consistency. With extensive experimental results, we demonstrate qualitatively and quantitatively that our model is able to deal with a large area of missing pixels in arbitrary shapes and generate realistic face completion results.

We present a transformation-grounded image generation network for novel 3D view synthesis from a single image. Our approach first explicitly infers the parts of the geometry visible both in the input and novel views and then casts the remaining synthesis problem as image completion. Specifically, we both predict a flow to move the pixels from the input to the novel view along with a novel visibility map that helps deal with occulsion/disocculsion. Next, conditioned on those intermediate results, we hallucinate (infer) parts of the object invisible in the input image. In addition to the new network structure, training with a combination of adversarial and perceptual loss results in a reduction in common artifacts of novel view synthesis such as distortions and holes, while successfully generating high frequency details and preserving visual aspects of the input image. We evaluate our approach on a wide range of synthetic and real examples. Both qualitative and quantitative results show our method achieves significantly better results compared to existing methods.

Video scene parsing is challenging due to the following two reasons: firstly, it is non-trivial to learn meaningful video representations for producing the temporally consistent labeling map; secondly, such a learning process becomes more difficult with insufficient labeled video training data. In this work, we propose a unified framework to address the above two problems, which is to our knowledge the first model to employ predictive feature learning in the video scene parsing. The predictive feature learning is carried out in two predictive tasks: frame prediction and predictive parsing. It is experimentally proved that the learned predictive features in our model are able to significantly enhance the video parsing performance by combining with the standard image parsing network. Interestingly, the performance gain brought by the predictive learning is almost costless as the features are learned from a large amount of unlabeled video data in an unsupervised way. Extensive experiments over two challenging datasets, Cityscapes and Camvid, have demonstrated the effectiveness of our model by showing remarkable improvement over well-established baselines.

Parsing urban scene images benefits many applications, especially self-driving. Most of the current solutions employ generic image parsing models that treat all scales and locations in the images equally and do not consider the geometry property of car-captured urban scene images. Thus, they suffer from heterogeneous object scales caused by perspective projection of cameras on actual scenes and inevitably encounter parsing failures on distant objects as well as other boundary and recognition errors. In this work, we propose a new FoveaNet model to fully exploit the perspective geometry of scene images and address the common failures of generic parsing models. FoveaNet estimates the perspective geometry of a scene image through a convolutional network which integrates supportive evidence from contextual objects within the image. Based on the perspective geometry information, FoveaNet “undoes” the camera perspective projection - analyzing regions in the space of the actual scene, and thus provides much more reliable parsing results. Furthermore, to effectively address the recognition errors, FoveaNet introduces a new dense CRFs model that takes the perspective geometry as a prior potential. We evaluate FoveaNet on two urban scene parsing datasets, Cityspaces and CamVid, which demonstrates that FoveaNet can outperform all the well-established baselines and provide new state-of-the-art performance.

The ability to edit materials of objects in images is desirable by many content creators. However, this is an extremely challenging task as it requires to disentangle intrinsic physical properties of an image. We propose an end-to-end network architecture that replicates the forward image formation process to accomplish this task. Specifically, given a single image, the network first predicts intrinsic properties, i.e. shape, illumination, and material, which are then provided to a rendering layer. This layer performs in-network image synthesis, thereby enabling the network to understand the physics behind the image formation process. The proposed rendering layer is fully differentiable, supports both diffuse and specular materials, and thus can be applicable in a variety of problem settings. We demonstrate a rich set of visually plausible material editing examples and provide an extensive comparative study.

In this paper we are interested in the problem of image segmentation given natural language descriptions, i.e. referring expressions. Existing works tackle this problem by first modeling images and sentences independently and then segment images by combining these two types of representations. We argue that learning word-to-image interaction is more native in the sense of jointly modeling two modalities for the image segmentation task, and we propose convolutional multimodal LSTM to encode the sequential interactions between individual words, visual information, and spatial information. We show that our proposed model outperforms the baseline model on benchmark datasets. In addition, we analyze the intermediate output of the proposed multimodal LSTM approach and empirically explain how this approach enforces a more effective word-to-image interaction.

The success of various applications including robotics, digital content creation, and visualization demand a structured and abstract representation of the 3D world from limited sensor data. Inspired by the nature of human perception of 3D shapes as a collection of simple parts, we explore such an abstract shape representation based on primitives. Given a single depth image of an object, we present 3DPRNN, a generative recurrent neural network that synthesizes multiple plausible shapes composed of a set of primitives. Our generative model encodes symmetry characteristics of common man-made objects, preserves long-range structural coherence, and describes objects of varying complexity with a compact representation. We also propose a method based on Gaussian Fields to generate a large scale dataset of primitive-based shape representations to train our network. We evaluate our approach on a wide range of examples and show that it outperforms nearest-neighbor based shape retrieval methods and is on-par with voxelbased generative models while using a significantly reduced parameter space.

We propose a hierarchical approach for making long-term predictions of future frames. To avoid inherent compounding errors in recursive pixel-level prediction, we propose to first estimate high-level structure in the input frames, then predict how that structure evolves in the future, and finally by observing a single frame from the past and the predicted high-level structure, we construct the future frames without having to observe any of the pixel-level predictions. Long-term video prediction is difficult to perform by recurrently observing the predicted frames because the small errors in pixel space exponentially amplify as predictions are made deeper into the future. Our approach prevents pixel-level error propagation from happening by removing the need to observe the predicted frames. Our model is built with a combination of LSTM and analogy based encoder-decoder convolutional neural networks, which independently predict the video structure and generate the future frames, respectively. In experiments, our model is evaluated on the Human3.6M and Penn Action datasets on the task of long-term pixel-level video prediction of humans performing actions and demonstrate significantly better results than the state-of-the-art.

It is important for intelligent systems, e.g. autonomous vehicles and robotics to anticipate the future in order to plan early and make decisions accordingly. Predicting the future scene parsing and motion dynamics helps the agents better understand the visual environment better as the former provides dense semantic segmentations, i.e. what objects will be present and where they will appear, while the latter provides dense motion information, i.e. how the objects move in the future. In this paper, we propose a novel model to predict the scene parsing and motion dynamics in unobserved future video frames simultaneously. Using history information (preceding frames and corresponding scene parsing results) as input, our model is able to predict the scene parsing and motion for arbitrary time steps ahead. More importantly, our model is superior compared to other methods that predict parsing and motion separately, as the complementary relationship between the two tasks are fully utilized in our model through joint learning. To our best knowledge, this is the first attempt in jointly predicting scene parsing and motion dynamics in the future frames. On the large-scale Cityscapes dataset, it is demonstrated that our model produces significantly better parsing and motion prediction results compared to well established baselines. In addition, we also show our model can be used to predict the steering angle of the vehicles, which further verifies the ability of our model to learn underlying latent parameters.

Universal style transfer aims to transfer arbitrary visual styles to content images. Existing feed-forward based methods, while enjoying the inference efficiency, are mainly limited by inability of generalizing to unseen styles or compromised visual quality. In this paper, we present a simple yet effective method that tackles these limitations without training on any pre-defined styles. The key ingredient of our method is a pair of feature transforms, whitening and coloring, that are embedded to an image reconstruction network. The whitening and coloring transforms reflect direct matching of feature covariance of the content image to a given style image, which shares similar spirits with the optimization of Gram matrix based cost in neural style transfer. We demonstrate the effectiveness of our algorithm by generating high-quality stylized images with comparisons to a number of recent methods. We also analyze our method by visualizing the whitened features and synthesizing textures by simple feature coloring.

Top-down visual saliency is an important module of visual attention. In this work, we propose a novel top-down saliency model that jointly learns a Conditional Random Field (CRF) and a visual dictionary. The proposed model incorporates a layered structure from top to bottom: CRF, sparse coding and image patches. With sparse coding as an intermediate layer, CRF is learned in a feature-adaptive manner; meanwhile with CRF as the output layer, the dictionary is learned under structured supervision. For efficient and effective joint learning, we develop a max-margin approach via a stochastic gradient descent algorithm. Experimental results on the Graz-02 and PASCAL VOC datasets show that our model performs favorably against state-of-the-art top-down saliency methods for target object localization. In addition, the dictionary update significantly improves the performance of our model. We demonstrate the merits of the proposed top-down saliency model by applying it to prioritizing object proposals for detection and predicting human fixations.

Structured support vector machine (SSVM) based methods have demonstrated encouraging performance in recent object tracking benchmarks. However, the complex and expensive optimization limits their deployment in real-world applications. In this paper, we present a simple yet efficient dual linear SSVM (DLSSVM) algorithm to enable fast learning and execution during tracking. By analyzing the dual variables, we propose a primal classifier update formula where the learning step size is computed in closed form. This online learning method significantly improves the robustness of the proposed linear SSVM with lower computational cost. Second, we approximate the intersection kernel for feature representations with an explicit feature map to further improve tracking performance. Finally, we extend the proposed DLSSVM tracker with multi-scale estimation to address the "drift" problem. Experimental results on large benchmark datasets with 50 and 100 video sequences show that the proposed DLSSVM tracking algorithm achieves state-of-the-art performance.

Interactive object selection is a very important research problem and has many applications. Previous algorithms require substantial user interactions to estimate the foreground and background distributions. In this paper, we present a novel deep-learning-based algorithm which has much better understanding of objectness and can reduce user interactions to just a few clicks. Our algorithm transforms user-provided positive and negative clicks into two Euclidean distance maps which are then concatenated with the RGB channels of images to compose (image, user interactions) pairs. We generate many of such pairs by combining several random sampling strategies to model users' click patterns and use them to finetune deep Fully Convolutional Networks (FCNs). Finally the output probability maps of our FCN-8s model is integrated with graph cut optimization to refine the boundary segments. Our model is trained on the PASCAL segmentation dataset and evaluated on other datasets with different object classes. Experimental results on both seen and unseen objects demonstrate that our algorithm has a good generalization ability and is superior to all existing interactive object selection approaches.

We develop a deep learning algorithm for contour detection with a fully convolutional encoder-decoder network. Different from previous low-level edge detection, our algorithm focuses on detecting higher-level object contours. Our network is trained end-to-end on PASCAL VOC with refined ground truth from inaccurate polygon annotations, yielding much higher precision in object contour detection than previous methods. We find that the learned model generalizes well to unseen object classes from the same supercategories on MS COCO and can match state-of-the-art edge detection on BSDS500 with fine-tuning. By combining with the multiscale combinatorial grouping algorithm, our method can generate high-quality segmented object proposals, which significantly advance the state-of-the-art on PASCAL VOC (improving average recall from 0.62 to 0.67) with a relatively small amount of candidates (~1660 per image).

Understanding the 3D world is a fundamental problem in computer vision. However, learning a good representation of 3D objects is still an open problem due to the high dimensionality of the data and many factors of variation involved. In this work, we investigate the task of single-view 3D object reconstruction from a learning agent's perspective. We formulate the learning process as an interaction between 3D and 2D representations and propose an encoder-decoder network with a novel projection loss defined by the projective transformation. More importantly, the projection loss enables the unsupervised learning using 2D observation without explicit 3D supervision. We demonstrate the ability of the model in generating 3D volume from a single 2D image with three sets of experiments: (1) learning from single-class objects; (2) learning from multi-class objects and (3) testing on novel object classes. Results show superior performance and better generalization ability for 3D object reconstruction when the projection loss is involved.

This paper formulates face labeling as a conditional random field with unary and pairwise classifiers. We develop a novel multi-objective learning method that optimizes a single unified deep convolutional network with two distinct non-structured loss functions: one encoding the unary label likelihoods and the other encoding the pairwise label dependencies. Moreover, we regularize the network by using a nonparametric prior as new input channels in addition to the RGB image, and show that significant performance improvements can be achieved with a much smaller network size. Experiments on both the LFW and Helen datasets demonstrate state-of-the-art results of the proposed algorithm, and accurate labeling results on challenging images can be obtained by the proposed algorithm for real-world applications.

Object segmentation is highly desirable for image understanding and editing. Current interactive tools require a great deal of user effort while automatic methods are usually limited to images of special object categories or with high color contrast. In this paper, we propose a data-driven algorithm that uses examples to break through these limits. As similar objects tend to share similar local shapes, we match query image patches with example images in multiscale to enable local shape transfer. The transferred local shape masks constitute a patch-level segmentation solution space and we thus develop a novel cascade algorithm, PatchCut, for coarse-to-fine object segmentation. In each stage of the cascade, local shape mask candidates are selected to refine the estimated segmentation of the previous stage iteratively with color models. Experimental results on various datasets (Weizmann Horse, Fashionista, Object Discovery and PASCAL) demonstrate the effectiveness and robustness of our algorithm.

An important problem for both graphics and vision is to synthesize novel views of a 3D object from a single image. This is in particular challenging due to the partial observability inherent in projecting a 3D object onto the image space, and the ill-posedness of inferring object shape and pose. However, we can train a neural network to address the problem if we restrict our attention to specific object classes (in our case faces and chairs) for which we can gather ample training data. In this paper, we propose a novel recurrent convolutional encoder-decoder network that is trained end-to-end on the task of rendering rotated objects starting from a single image. The recurrent structure allows our model to capture long- term dependencies along a sequence of transformations, and we demonstrate the quality of its predictions for human faces on the Multi-PIE dataset and for a dataset of 3D chair models, and also show its ability of disentangling latent data factors without using object class labels.

We present two approaches to extract regions from structured edge detection. While the state-of-the-art algorithm based on globalized probability of boundary (gPb) generates a hierarchical region tree, it entails significant computational load. In this work, we exploit an efficient algorithm for structured edge prediction to extract regions. To generate high quality regions, we develop a novel algorithm to link the structured edge and gPb hierarchical image segmentation framework with steerable filters. The extracted regions are grouped by the proposed hierarchical grouping method to generate object proposals for effective detection and recognition problems. We demonstrate the effectiveness of our region generation for image segmentation on the BSDS500 database, and region generation for object proposals on the PASCAL VOC 2007 benchmark database. Experimental results show that the proposed algorithm achieves the comparable or superior quality to the state-of-the-art methods.

Appearance-based human re-identification is challenging due to different camera characteristics, varying lighting conditions, pose variations across camera views, etc. Recent studies have revealed that color information plays a critical role on performance. However, two problems remain unclear: (1) how do different color descriptors perform under the same scene in re-identification problem? and (2) how can we combine these descriptors without losing their invariance property and distinctiveness power? In this paper, we propose a novel ensemble model that combines different color descriptors in the decision level through metric learning. Experiments show that the proposed system significantly outperforms state-of-the-art algorithms on two challenging datasets (VIPeR and PRID 450S). We have improved the Rank 1 recognition rate on VIPeR dataset by 8.7%.

This paper presents a scalable scene parsing algorithm based on image retrieval and superpixel matching. We focus on rare object classes, which play an important role in achieving richer semantic understanding of visual scenes, compared to common background classes. Towards this end, we make two novel contributions: rare class expansion and semantic context description. First, considering the long-tailed nature of the label distribution, we expand the retrieval set by rare class exemplars and thus achieve more balanced superpixel classification results. Second, we incorporate both global and local semantic context information through a feedback based mechanism to refine image retrieval and superpixel matching. Results on the SIFTflow and LMSun datasets show the superior performance of our algorithm, especially on the rare classes, without sacrificing overall labeling accuracy.

We present Max-Margin Boltzmann Machines (MMBMs) for object segmentation. MMBMs are essentially a class of Conditional Boltzmann Machines that model the joint distribution of hidden variables and output labels conditioned on input observations. In addition to image-to-label connections, we build direct image-to-hidden connections to facilitate global shape prediction, and thus derive a simple Iterated Conditional Modes algorithm for efficient maximum a posteriori inference. We formulate a max-margin objective function for discriminative training, and analyze the effects of different margin functions on learning. We evaluate MMBMs using three datasets against state-of-the-art methods to demonstrate the strength of the proposed algorithms.

We present a hybrid parametric and nonparametric algorithm, exemplar cut, for generating class-specific object segmentation hypotheses. For the parametric part, we train a pylon model on a hierarchical region tree as the energy function for segmentation. For the nonparametric part, we match the input image with each exemplar by using regions to obtain a score which augments the energy function from the pylon model. Our method thus generates a set of highly plausible segmentation hypotheses by solving a series of exemplar augmented graph cuts. Experimental results on the Graz and PASCAL datasets show that the proposed algorithm achieves favorable segmentation performance against the state-of-the-art methods in terms of visual quality and accuracy.

Top-down visual saliency facilities object localization by providing a discriminative representation of target objects and a probability map for reducing the search space. In this paper, we propose a novel top-down saliency model that jointly learns a Conditional Random Field (CRF) and a discriminative dictionary. The proposed model is formulated based on a CRF with latent variables. By using sparse codes as latent variables, we train the dictionary modulated by CRF, and meanwhile a CRF with sparse coding. We propose a max-margin approach to train our model via fast inference algorithms. We evaluate our model on the Graz-02 and PASCAL VOC 2007 datasets. Experimental results show that our model performs favorably against the state-of-the-art top-down saliency methods. We also observe that the dictionary update significantly improves the model performance.

Visual prior from generic real-world images can be learned and transferred for representing objects in a scene. Motivated by this, we propose an algorithm that transfers visual prior learned offline for online object tracking. From a collection of real-world images, we learn an overcomplete dictionary to represent visual prior. The prior knowledge of objects is generic, and the training image set does not necessarily contain any observation of the target object. During the tracking process, the learned visual prior is transferred to construct an object representation by sparse coding and multiscale max pooling. With this representation, a linear classifier is learned online to distinguish the target from the background and to account for the target and background appearance variations over time. Tracking is then carried out within a Bayesian inference framework, in which the learned classifier is used to construct the observation model and a particle filter is used to estimate the tracking result sequentially. Experiments on a variety of challenging sequences with comparisons to several state-of-the-art methods demonstrate that more robust object tracking can be achieved by transferring visual prior.

This paper presents a deep learning model of building up hierarchical image representation. Each layer of hierarchy consists of three components: sparse coding, saliency pooling and local grouping. With sparse coding we identify distinctive coefficients for representing raw features of each lower layer; saliency pooling helps suppress noise and enhance translation invariance of sparse representation; we group locally pooled sparse codes to form more complex representations. Instead of using hand-crafted descriptors, our model learns an effective image representation directly from images in a unsupervised data-driven manner. We evaluate our algorithm with several benchmark databases of object recognition and analyze the contributions of different components. Experimental results show that our algorithm performs favorably against the state-of-the-art methods.

This paper deals with a new problem in face recognition research, in which the enrollment and query face samples are captured under different lighting conditions. In our case, the enrollment samples are visual light (VIS) images, whereas the query samples are taken under near infrared (NIR) condition. It is very difficult to directly match the face samples captured under these two lighting conditions due to their different visual appearances. In this paper, we propose a novel method for synthesizing VIS images from NIR images based on learning the mappings between images of different spectra (i.e., NIR and VIS). In our approach, we reduce the inter-spectral differences significantly, thus allowing effective matching between faces taken under different imaging conditions. Face recognition experiments clearly show the efficacy of the proposed approach.