Generating speech-consistent body and gesture movements is a long-standing problem in virtual avatar creation. Previous studies often synthesize pose movement in a holistic manner, where poses of all joints are generated simultaneously. Such a straightforward pipeline fails to generate fine-grained co-speech gestures. One observation is that the hierarchical semantics in speech and the hierarchical structures of human gestures can be naturally described into multiple granularities and associated together. To fully utilize the rich connections between speech audio and human gestures, we propose a novel framework named Hierarchical Audio-to-Gesture (HA2G) for co-speech gesture generation. In HA2G, a Hierarchical Audio Learner extracts audio representations across semantic granularities. A Hierarchical Pose Inferer subsequently renders the entire human pose gradually in a hierarchical manner. To enhance the quality of synthesized gestures, we develop a contrastive learning strategy based on audio-text alignment for better audio representations. Extensive experiments and human evaluation demonstrate that the proposed method renders realistic co-speech gestures and outperforms previous methods in a clear margin. Project page: https://alvinliu0.github.io/projects/HA2G
Understanding a person's behavior from their 3D motion is a fundamental problem in computer vision with many applications. An important component of this problem is 3D Temporal Action Localization (3D-TAL), which involves recognizing what actions a person is performing, and when. State-of-the-art 3D-TAL methods employ a two-stage approach in which the action span detection task and the action recognition task are implemented as a cascade. This approach, however, limits the possibility of error-correction. In contrast, we propose LocATe, an end-to-end approach that jointly localizes and recognizes actions in a 3D sequence. Further, unlike existing autoregressive models that focus on modeling the local context in a sequence, LocATe's transformer model is capable of capturing long-term correlations between actions in a sequence. Unlike transformer-based object-detection and classification models which consider image or patch features as input, the input in 3D-TAL is a long sequence of highly correlated frames. To handle the high-dimensional input, we implement an effective input representation, and overcome the diffuse attention across long time horizons by introducing sparse attention in the model. LocATe outperforms previous approaches on the existing PKU-MMD 3D-TAL benchmark (mAP=93.2%). Finally, we argue that benchmark datasets are most useful where there is clear room for performance improvement. To that end, we introduce a new, challenging, and more realistic benchmark dataset, BABEL-TAL-20 (BT20), where the performance of state-of-the-art methods is significantly worse. The dataset and code for the method will be available for research purposes.
Human intervention is an effective way to inject human knowledge into the training loop of reinforcement learning, which can bring fast learning and ensured training safety. Given the very limited budget of human intervention, it remains challenging to design when and how human expert interacts with the learning agent in the training. In this work, we develop a novel human-in-the-loop learning method called Human-AI Copilot Optimization (HACO).To allow the agent's sufficient exploration in the risky environments while ensuring the training safety, the human expert can take over the control and demonstrate how to avoid probably dangerous situations or trivial behaviors. The proposed HACO then effectively utilizes the data both from the trial-and-error exploration and human's partial demonstration to train a high-performing agent. HACO extracts proxy state-action values from partial human demonstration and optimizes the agent to improve the proxy values meanwhile reduce the human interventions. The experiments show that HACO achieves a substantially high sample efficiency in the safe driving benchmark. HACO can train agents to drive in unseen traffic scenarios with a handful of human intervention budget and achieve high safety and generalizability, outperforming both reinforcement learning and imitation learning baselines with a large margin. Code and demo videos are available at: https://decisionforce.github.io/HACO/.
The task of audio-visual sound source localization has been well studied under constrained scenes, where the audio recordings are clean. However, in real-world scenarios, audios are usually contaminated by off-screen sound and background noise. They will interfere with the procedure of identifying desired sources and building visual-sound connections, making previous studies non-applicable. In this work, we propose the Interference Eraser (IEr) framework, which tackles the problem of audio-visual sound source localization in the wild. The key idea is to eliminate the interference by redefining and carving discriminative audio representations. Specifically, we observe that the previous practice of learning only a single audio representation is insufficient due to the additive nature of audio signals. We thus extend the audio representation with our Audio-Instance-Identifier module, which clearly distinguishes sounding instances when audio signals of different volumes are unevenly mixed. Then we erase the influence of the audible but off-screen sounds and the silent but visible objects by a Cross-modal Referrer module with cross-modality distillation. Quantitative and qualitative evaluations demonstrate that our proposed framework achieves superior results on sound localization tasks, especially under real-world scenarios. Code is available at https://github.com/alvinliu0/Visual-Sound-Localization-in-the-Wild.
Animating high-fidelity video portrait with speech audio is crucial for virtual reality and digital entertainment. While most previous studies rely on accurate explicit structural information, recent works explore the implicit scene representation of Neural Radiance Fields (NeRF) for realistic generation. In order to capture the inconsistent motions as well as the semantic difference between human head and torso, some work models them via two individual sets of NeRF, leading to unnatural results. In this work, we propose Semantic-aware Speaking Portrait NeRF (SSP-NeRF), which creates delicate audio-driven portraits using one unified set of NeRF. The proposed model can handle the detailed local facial semantics and the global head-torso relationship through two semantic-aware modules. Specifically, we first propose a Semantic-Aware Dynamic Ray Sampling module with an additional parsing branch that facilitates audio-driven volume rendering. Moreover, to enable portrait rendering in one unified neural radiance field, a Torso Deformation module is designed to stabilize the large-scale non-rigid torso motions. Extensive evaluations demonstrate that our proposed approach renders more realistic video portraits compared to previous methods. Project page: https://alvinliu0.github.io/projects/SSP-NeRF
Object detection through either RGB images or the LiDAR point clouds has been extensively explored in autonomous driving. However, it remains challenging to make these two data sources complementary and beneficial to each other. In this paper, we propose \textit{AutoAlign}, an automatic feature fusion strategy for 3D object detection. Instead of establishing deterministic correspondence with camera projection matrix, we model the mapping relationship between the image and point clouds with a learnable alignment map. This map enables our model to automate the alignment of non-homogenous features in a dynamic and data-driven manner. Specifically, a cross-attention feature alignment module is devised to adaptively aggregate \textit{pixel-level} image features for each voxel. To enhance the semantic consistency during feature alignment, we also design a self-supervised cross-modal feature interaction module, through which the model can learn feature aggregation with \textit{instance-level} feature guidance. Extensive experimental results show that our approach can lead to 2.3 mAP and 7.0 mAP improvements on the KITTI and nuScenes datasets, respectively. Notably, our best model reaches 70.9 NDS on the nuScenes testing leaderboard, achieving competitive performance among various state-of-the-arts.
Making generative models 3D-aware bridges the 2D image space and the 3D physical world yet remains challenging. Recent attempts equip a Generative Adversarial Network (GAN) with a Neural Radiance Field (NeRF), which maps 3D coordinates to pixel values, as a 3D prior. However, the implicit function in NeRF has a very local receptive field, making the generator hard to become aware of the global structure. Meanwhile, NeRF is built on volume rendering which can be too costly to produce high-resolution results, increasing the optimization difficulty. To alleviate these two problems, we propose a novel framework, termed as VolumeGAN, for high-fidelity 3D-aware image synthesis, through explicitly learning a structural representation and a textural representation. We first learn a feature volume to represent the underlying structure, which is then converted to a feature field using a NeRF-like model. The feature field is further accumulated into a 2D feature map as the textural representation, followed by a neural renderer for appearance synthesis. Such a design enables independent control of the shape and the appearance. Extensive experiments on a wide range of datasets show that our approach achieves sufficiently higher image quality and better 3D control than the previous methods.
Semi-supervised action recognition is a challenging but important task due to the high cost of data annotation. A common approach to this problem is to assign unlabeled data with pseudo-labels, which are then used as additional supervision in training. Typically in recent work, the pseudo-labels are obtained by training a model on the labeled data, and then using confident predictions from the model to teach itself. In this work, we propose a more effective pseudo-labeling scheme, called Cross-Model Pseudo-Labeling (CMPL). Concretely, we introduce a lightweight auxiliary network in addition to the primary backbone, and ask them to predict pseudo-labels for each other. We observe that, due to their different structural biases, these two models tend to learn complementary representations from the same video clips. Each model can thus benefit from its counterpart by utilizing cross-model predictions as supervision. Experiments on different data partition protocols demonstrate the significant improvement of our framework over existing alternatives. For example, CMPL achieves $17.6\%$ and $25.1\%$ Top-1 accuracy on Kinetics-400 and UCF-101 using only the RGB modality and $1\%$ labeled data, outperforming our baseline model, FixMatch, by $9.0\%$ and $10.3\%$, respectively.
Pre-training has become a standard paradigm in many computer vision tasks. However, most of the methods are generally designed on the RGB image domain. Due to the discrepancy between the two-dimensional image plane and the three-dimensional space, such pre-trained models fail to perceive spatial information and serve as sub-optimal solutions for 3D-related tasks. To bridge this gap, we aim to learn a spatial-aware visual representation that can describe the three-dimensional space and is more suitable and effective for these tasks. To leverage point clouds, which are much more superior in providing spatial information compared to images, we propose a simple yet effective 2D Image and 3D Point cloud Unsupervised pre-training strategy, called SimIPU. Specifically, we develop a multi-modal contrastive learning framework that consists of an intra-modal spatial perception module to learn a spatial-aware representation from point clouds and an inter-modal feature interaction module to transfer the capability of perceiving spatial information from the point cloud encoder to the image encoder, respectively. Positive pairs for contrastive losses are established by the matching algorithm and the projection matrix. The whole framework is trained in an unsupervised end-to-end fashion. To the best of our knowledge, this is the first study to explore contrastive learning pre-training strategies for outdoor multi-modal datasets, containing paired camera images and LIDAR point clouds. Codes and models are available at https://github.com/zhyever/SimIPU.
The success of Generative Adversarial Networks (GANs) is largely built upon the adversarial training between a generator (G) and a discriminator (D). They are expected to reach a certain equilibrium where D cannot distinguish the generated images from the real ones. However, in practice it is difficult to achieve such an equilibrium in GAN training, instead, D almost always surpasses G. We attribute this phenomenon to the information asymmetry between D and G. Specifically, we observe that D learns its own visual attention when determining whether an image is real or fake, but G has no explicit clue on which regions to focus on for a particular synthesis. To alleviate the issue of D dominating the competition in GANs, we aim to raise the spatial awareness of G. Randomly sampled multi-level heatmaps are encoded into the intermediate layers of G as an inductive bias. Thus G can purposefully improve the synthesis of certain image regions. We further propose to align the spatial awareness of G with the attention map induced from D. Through this way we effectively lessen the information gap between D and G. Extensive results show that our method pushes the two-player game in GANs closer to the equilibrium, leading to a better synthesis performance. As a byproduct, the introduced spatial awareness facilitates interactive editing over the output synthesis. Demo video and more results are at https://genforce.github.io/eqgan/.