CVPR 2022 Papers

Speech-driven 3D facial animation is challenging due to the complex geometry of human faces and the limited availability of 3D audio-visual data. Prior works typically focus on learning phoneme-level features of short audio windows with limited context, occasionally resulting in inaccurate lip movements. To tackle this limitation, we propose a Transformer-based autoregressive model, FaceFormer, which encodes the long-term audio context and autoregressively predicts a sequence of animated 3D face meshes. To cope with the data scarcity issue, we integrate the self-supervised pre-trained speech representations. Also, we devise two biased attention mechanisms well suited to this specific task, including the biased cross-modal multi-head (MH) attention and the biased causal MH self-attention with a periodic positional encoding strategy. The former effectively aligns the audio-motion modalities, whereas the latter offers abilities to generalize to longer audio sequences. Extensive experiments and a perceptual user study show that our approach outperforms the existing state-of-the-arts. The code and the video are available at: https://evelynfan.github.io/audio2face/.

Facial expression recognition (FER) is challenging due to the class imbalance caused by data collection. Existing studies tackle the data bias problem using only labeled facial expression dataset. Orthogonal to existing FER methods, we propose to utilize large unlabeled face recognition (FR) datasets to enhance FER. However, this raises another data bias problem---the distribution mismatch between FR and FER data. To combat the mismatch, we propose the Meta-Face2Exp framework, which consists of a base network and an adaptation network. The base network learns prior expression knowledge on class-balanced FER data while the adaptation network is trained to fit the pseudo labels of FR data generated by the base model. To combat the mismatch between FR and FER data, Meta-Face2Exp uses a circuit feedback mechanism, which improves the base network with the feedback from the adaptation network. Experiments show that our Meta-Face2Exp achieves comparable accuracy to state-of-the-art FER methods with 10% of the labeled FER data utilized by the baselines. We also demonstrate that the circuit feedback mechanism successfully eliminates data bias.

Most face relighting methods are able to handle diffuse shadows, but struggle to handle hard shadows, such as those cast by the nose. Methods that propose techniques for handling hard shadows often do not produce geometrically consistent shadows since they do not directly leverage the estimated face geometry while synthesizing them. We propose a novel differentiable algorithm for synthesizing hard shadows based on ray tracing, which we incorporate into training our face relighting model. Our proposed algorithm directly utilizes the estimated face geometry to synthesize geometrically consistent hard shadows. We demonstrate through quantitative and qualitative experiments on Multi-PIE and FFHQ that our method produces more geometrically consistent shadows than previous face relighting methods while also achieving state-of-the-art face relighting performance under directional lighting. In addition, we demonstrate that our differentiable hard shadow modeling improves the quality of the estimated face geometry over diffuse shading models.

Novel view synthesis (NVS) is a challenging task requiring systems to generate photorealistic images of scenes from new viewpoints, where both quality and speed are important for applications. Previous image-based rendering (IBR) methods are fast, but have poor quality when input views are sparse. Recent Neural Radiance Fields (NeRF) and generalizable variants give impressive results but are not real-time. In our paper, we propose a generalizable NVS method with sparse inputs, called \FWDds, which gives high-quality synthesis in real-time. With explicit depth and differentiable rendering, it achieves competitive results to the SOTA methods with 130-1000xspeedup and better perceptual quality. If available, we can seamlessly integrate sensor depth during either training or inference to improve image quality while retaining real-time speed. With the growing prevalence of depths sensors, we hope that methods making use of depth will become increasingly useful.

6D object pose estimation networks are limited in their capability to scale to large numbers of object instances due to the close-set assumption and their reliance on high-fidelity object CAD models. In this work, we study a new open set problem; the few-shot 6D object poses estimation: estimating the 6D pose of an unknown object by a few support views without extra training. To tackle the problem, we point out the importance of fully exploring the appearance and geometric relationship between the given support views and query scene patches and propose a dense prototypes matching framework by extracting and matching dense RGBD prototypes with transformers. Moreover, we show that the priors from diverse appearances and shapes are crucial to the generalization capability under the problem setting and thus propose a large-scale RGBD photorealistic dataset (ShapeNet6D) for network pre-training. A simple and effective online texture blending approach is also introduced to eliminate the domain gap from the synthesis dataset, which enriches appearance diversity at a low cost. Finally, we discuss possible solutions to this problem and establish benchmarks on popular datasets to facilitate future research.

For Visible-Infrared person Re-IDentification (VI-ReID), existing modality-specific information compensation based models try to generate the images of missing modality from existing ones for reducing cross-modality discrepancy. However, because of the large modality discrepancy between visible and infrared images, the generated images usually have low qualities and introduce much more interfering information (e.g., color inconsistency). This greatly degrades the subsequent VI-ReID performance. Alternatively, we present a novel Feature-level Modality Compensation Network (FMCNet) for VIReID in this paper, which aims to compensate the missing modality-specific information in the feature level rather than in the image level, i.e., directly generating those missing modality-specific features of one modality from existing modality-shared features of the other modality. This will enable our model to mainly generate some discriminative person related modality-specific features and discard those non-discriminative ones for benefiting VI-ReID. For that, a single-modality feature decomposition module is first designed to decompose single-modality features into modality-specific ones and modality-shared ones. Then, a feature-level modality compensation module is present to generate those missing modality-specific features from existing modality-shared ones. Finally, a shared-specific feature fusion module is proposed to combine the existing and generated features for VI-ReID. The effectiveness of our proposed model is verified on two benchmark datasets.

Multi-object multi-part scene parsing is a challenging task which requires detecting multiple object classes in a scene and segmenting the semantic parts within each object. In this paper, we propose FLOAT, a factorized label space framework for scalable multi-object multi-part parsing. Our framework involves independent dense prediction of object category and part attributes which increases scalability and reduces task complexity compared to the monolithic label space counterpart. In addition, we propose an inference-time 'zoom' refinement technique which significantly improves segmentation quality, especially for smaller objects/parts. Compared to state of the art, FLOAT obtains an absolute improvement of 2.0% for mean IOU (mIOU) and 4.8% for segmentation quality IOU (sqIOU) on the Pascal-Part-58 dataset. For the larger Pascal-Part-108 dataset, the improvements are 2.1% for mIOU and 3.9% for sqIOU. We incorporate previously excluded part attributes and other minor parts of the Pascal-Part dataset to create the most comprehensive and challenging version which we dub Pascal-Part-201. FLOAT obtains improvements of 8.6% for mIOU and 7.5% for sqIOU on the new dataset, demonstrating its parsing effectiveness across a challenging diversity of objects and parts. The code and datasets are available at floatseg.github.io.

State-of-the-art vision and vision-and-language models rely on large-scale visio-linguistic pretraining for obtaining good performance on a variety of downstream tasks. Generally, such models are often either cross-modal (contrastive) or multi-modal (with earlier fusion) but not both; and they often only target specific modalities or tasks. A promising direction would be to use a single holistic universal model, as a "foundation", that targets all modalities at once---a true vision and language foundation model should be good at vision tasks, language tasks, and cross- and multi-modal vision and language tasks. We introduce FLAVA as such a model and demonstrate impressive performance on a wide range of 35 tasks spanning these target modalities.

To represent people in mixed reality applications for collaboration and communication, we need to generate realistic and faithful avatar poses. However, the signal streams that can be applied for this task from head-mounted devices (HMDs) are typically limited to head pose and hand pose estimates. While these signals are valuable, they are an incomplete representation of the human body, making it challenging to generate a faithful full-body avatar. We address this challenge by developing a flow-based generative model of the 3D human body from sparse observations, wherein we learn not only a conditional distribution of 3D human pose, but also a probabilistic mapping from observations to the latent space from which we can generate a plausible pose along with uncertainty estimates for the joints. We show that our approach is not only a strong predictive model, but can also act as an efficient pose prior in different optimization settings where a good initial latent code plays a major role.

Robust visual recognition under adverse weather conditions is of great importance in real-world applications. In this context, we propose a new method for learning semantic segmentation models robust against fog. Its key idea is to consider the fog condition of an image as its style and close the gap between images with different fog conditions in neural style spaces of a segmentation model. In particular, since the neural style of an image is in general affected by other factors as well as fog, we introduce a fog-pass filter module that learns to extract a fog-relevant factor from the style. Optimizing the fog-pass filter and the segmentation model alternately gradually closes the style gap between different fog conditions and allows to learn fog-invariant features in consequence. Our method substantially outperforms previous work on three real foggy image datasets. Moreover, it improves performance on both foggy and clear weather images, while existing methods often degrade performance on clear scenes.

In recent years, the security of AI systems has drawn increasing research attention, especially in the medical imaging realm. To develop a secure medical image analysis (MIA) system, it is a must to study possible backdoor attacks (BAs), which can embed hidden malicious behaviors into the system. However, designing a unified BA method that can be applied to various MIA systems is challenging due to the diversity of imaging modalities (e.g., X-Ray, CT, and MRI) and analysis tasks (e.g., classification, detection, and segmentation). Most existing BA methods are designed to attack natural image classification models, which apply spatial triggers to training images and inevitably corrupt the semantics of poisoned pixels, leading to the failures of attacking dense prediction models. To address this issue, we propose a novel Frequency-Injection based Backdoor Attack method (FIBA) that is capable of delivering attacks in various MIA tasks. Specifically, FIBA leverages a trigger function in the frequency domain that can inject the low-frequency information of a trigger image into the poisoned image by linearly combining the spectral amplitude of both images. Since it preserves the semantics of the poisoned image pixels, FIBA can perform attacks on both classification and dense prediction models. Experiments on three benchmarks in MIA (i.e., ISIC-2019 for skin lesion classification, KiTS-19 for kidney tumor segmentation, and EAD-2019 for endoscopic artifact detection), validate the effectiveness of FIBA and its superiority over state-of-the-art methods in attacking MIA models as well as bypassing backdoor defense. The code will be released.

Current benchmarks for facial expression recognition (FER) mainly focus on static images, while there are limited datasets for FER in videos. It is still ambiguous to evaluate whether performances of existing methods remain satisfactory in real-world application-oriented scenes. For example, the "Happy" expression with high intensity in Talk-Show is more discriminating than the same expression with low intensity in Official-Event. To fill this gap, we build a large-scale multi-scene dataset, coined as FERV39k. We analyze the important ingredients of constructing such a novel dataset in three aspects: (1) multi-scene hierarchy and expression class, (2) generation of candidate video clips, (3) trusted manual labelling process. Based on these guidelines, we select 4 scenarios subdivided into 22 scenes, annotate 86k samples automatically obtained from 4k videos based on the well-designed workflow, and finally build 38,935 video clips labeled with 7 classic expressions. Experiment benchmarks on four kinds of baseline frameworks were also provided and further analysis on their performance across different scenes and some challenges for future research were given. Besides, we systematically investigate key components of DFER by ablation studies. The baseline framework and our project are available on https://github.com/wangyanckxx/FERV39k.

Previous portrait image generation methods roughly fall into two categories: 2D GANs and 3D-aware GANs. 2D GANs can generate high fidelity portraits but with low view consistency. 3D-aware GAN methods can maintain view consistency but their generated images are not locally editable. To overcome these limitations, we propose FENeRF, a 3D-aware generator that can produce view-consistent and locally-editable portrait images. Our method uses two decoupled latent codes to generate corresponding facial semantics and texture in a spatial-aligned 3D volume with shared geometry. Benefiting from such underlying 3D representation, FENeRF can jointly render the boundary-aligned image and semantic mask and use the semantic mask to edit the 3D volume via GAN inversion. We further show such 3D representation can be learned from widely available monocular image and semantic mask pairs. Moreover, we reveal that joint learning semantics and texture helps to generate finer geometry. Our experiments demonstrate that FENeRF outperforms state-of-the-art methods in various face editing tasks.

In recent years, increasing attention has been drawn to the internal mechanisms of representation models. Traditional methods are inapplicable to fully explain the feature representations, especially if the images do not fit into any category. In this case, employing an existing class or the similarity with other image is unable to provide a complete and reliable visual explanation. To handle this task, we propose a novel visual explanation paradigm called Feature Activation Mapping (FAM) in this paper. Under this paradigm, Grad-FAM and Score-FAM are designed for visualizing feature representations. Unlike the previous approaches, FAM locates the regions of images that contribute most to the feature vector itself. Extensive experiments and evaluations, both subjective and objective, showed that Score-FAM provided most promising interpretable visual explanations for feature representations in Person Re-Identification. Furthermore, FAM also can be employed to analyze other vision tasks, such as self-supervised representation learning.

A practical eye authentication (EA) system targeted for edge devices needs to perform authentication and be robust to presentation attacks, all while remaining compute and latency efficient. However, existing eye-based frameworks a) perform authentication and Presentation Attack Detection (PAD) independently and b) involve significant pre-processing steps to extract the iris region. Here, we introduce a joint framework for EA and PAD using periocular images. While a deep Multitask Learning (MTL) network can perform both the tasks, MTL suffers from the forgetting effect since the training datasets for EA and PAD are disjoint. To overcome this, we propose Eye Authentication with PAD (EyePAD), a distillation-based method that trains a single network for EA and PAD while reducing the effect of forgetting. To further improve the EA performance, we introduce a novel approach called EyePAD++ that includes training an MTL network on both EA and PAD data, while distilling the 'versatility' of the EyePAD network through an additional distillation step. Our proposed methods outperform the SOTA in PAD and obtain near-SOTA performance in eye-to-eye verification, without any pre-processing. We also demonstrate the efficacy of EyePAD and EyePAD++ in user-to-user verification with PAD across network backbones and image quality.

We present an efficient method for joint optimization of topology, materials and lighting from multi-view image observations. Unlike recent multi-view reconstruction approaches, which typically produce entangled 3D representations encoded in neural networks, we output triangle meshes with spatially-varying materials and environment lighting that can be deployed in any traditional graphics engine unmodified. We leverage recent work in differentiable rendering, coordinate-based networks to compactly represent volumetric texturing, alongside differentiable marching tetrahedrons to enable gradient-based optimization directly on the surface mesh. Finally, we introduce a differentiable formulation of the split sum approximation of environment lighting to efficiently recover all-frequency lighting. Experiments show our extracted models used in advanced scene editing, material decomposition, and high quality view interpolation, all running at interactive rates in triangle-based renderers (rasterizers and path tracers).

Generating expressive talking heads is essential for creating virtual humans. However, existing one- or few-shot methods focus on lip-sync and head motion, ignoring the emotional expressions that make talking faces realistic. In this paper, we propose the Granularly Controlled Audio-Visual Talking Heads (GC-AVT), which controls lip movements, head poses, and facial expressions of a talking head in a granular manner. Our insight is to decouple the audio-visual driving sources through prior-based pre-processing designs. Detailedly, we disassemble the driving image into three complementary parts including: 1) a cropped mouth that facilitates lip-sync; 2) a masked head that implicitly learns pose; and 3) the upper face which works corporately and complementarily with a time-shifted mouth to contribute the expression. Interestingly, the encoded features from the three sources are integrally balanced through reconstruction training. Extensive experiments show that our method generates expressive faces with not only synced mouth shapes, controllable poses, but precisely animated emotional expressions as well.

Images captured with improper exposures usually bring unsatisfactory visual effects. Previous works mainly focus on either underexposure or overexposure correction, resulting in poor generalization to various exposures. An alternative solution is to mix the multiple exposure data for training a single network. However, the procedures of correcting underexposure and overexposure to normal exposures are much different from each other, leading to large discrepancies for the network in correcting multiple exposures, thus resulting in poor performance. The key point to address this issue lies in bridging different exposure representations. To achieve this goal, we design a multiple exposure correction framework based on an Exposure Normalization and Compensation (ENC) module. Specifically, the ENC module consists of an exposure normalization part for mapping different exposure features to the exposure-invariant feature space, and a compensation part for integrating the initial features unprocessed by exposure normalization part to ensure the completeness of information. Besides, to further alleviate the imbalanced performance caused by variations in the optimization process, we introduce a parameter regularization fine-tuning strategy to improve the performance of the worst-performed exposure without degrading other exposures. Our model empowered by ENC outperforms the existing methods by more than 2dB and is robust to multiple image enhancement tasks, demonstrating its effectiveness and generalization capability for real-world applications.

Self-supervised learning has shown its great potential to extract powerful visual representations without human annotations. Various works are proposed to deal with self-supervised learning from different perspectives: (1) contrastive learning methods (e.g., MoCo, SimCLR) utilize both positive and negative samples to guide the training direction; (2) asymmetric network methods (e.g., BYOL, SimSiam) get rid of negative samples via the introduction of a predictor network and the stop-gradient operation; (3) feature decorrelation methods (e.g., Barlow Twins, VICReg) instead aim to reduce the redundancy between feature dimensions. These methods appear to be quite different in the designed loss functions from various motivations. The final accuracy numbers also vary, where different networks and tricks are utilized in different works. In this work, we demonstrate that these methods can be unified into the same form. Instead of comparing their loss functions, we derive a unified formula through gradient analysis. Furthermore, we conduct fair and detailed experiments to compare their performances. It turns out that there is little gap between these methods, and the use of momentum encoder is the key factor to boost performance. From this unified framework, we propose UniGrad, a simple but effective gradient form for self-supervised learning. It does not require a memory bank or a predictor network, but can still achieve state-of-the-art performance and easily adopt other training strategies. Extensive experiments on linear evaluation and many downstream tasks also show its effectiveness. Code shall be released.