Generating accurate descriptions of human actions in videos remains a challenging task for video captioning models. Existing approaches often struggle to capture fine-grained motion details, resulting in vague or semantically inconsistent captions. In this work, we introduce the Motion-Augmented Caption Model (M-ACM), a novel generative framework that enhances caption quality by incorporating motion-aware decoding. At its core, M-ACM leverages motion representations derived from human mesh recovery to explicitly highlight human body dynamics, thereby reducing hallucinations and improving both semantic fidelity and spatial alignment in the generated captions. To support research in this area, we present the Human Motion Insight (HMI) Dataset, comprising 115K video-description pairs focused on human movement, along with HMI-Bench, a dedicated benchmark for evaluating motion-focused video captioning. Experimental results demonstrate that M-ACM significantly outperforms previous methods in accurately describing complex human motions and subtle temporal variations, setting a new standard for motion-centric video captioning.

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

Semantic segmentation has recently achieved notable advances by exploiting “class-level” contextual information during learning, e.g., the Object Contextual Representation (OCR) and Context Prior (CPNet) approaches. However, these approaches simply concatenate class-level information to pixel features to boost pixel representation learning, which cannot fully utilize intra-class and inter-class contextual information. Moreover, these approaches learn soft class centers based on coarse mask prediction, which is prone to error accumulation. To better exploit class-level information, we propose a universal Class-Aware Regularization (CAR) approach to optimize the intra-class variance and inter-class distance during feature learning, motivated by the fact that humans can recognize an object by itself no matter which other objects it appears with. Moreover, we design a dedicated decoder for CAR (named CARD), which consists of a novel spatial token mixer and an upsampling module, to maximize its gain for existing baselines while being highly efficient in terms of computational cost. Specifically, CAR consists of three novel loss functions. The first loss function encourages more compact class representations within each class, the second directly maximizes the distance between different class centers, and the third further pushes the distance between inter-class centers and pixels. Furthermore, the class center in our approach is directly generated from ground truth instead of from the error-prone coarse prediction. CAR can be directly applied to most existing segmentation models during training, including OCR and CPNet, and can largely improve their accuracy at no additional inference overhead. Extensive experiments and ablation studies conducted on multiple benchmark datasets demonstrate that the proposed CAR can boost the accuracy of all baseline models by up to 2.23% mIOU with superior generalization ability. CARD outperforms state-of-the-art approaches on multiple benchmarks with a highly efficient architecture. The code will be available at https://github.com/edwardyehuang/CAR.

People may perform diverse gestures affected by various mental and physical factors when speaking the same sentences. This inherent one-to-many relationship makes co-speech gesture generation from audio particularly challenging. Conventional CNNs/RNNs assume one-to-one mapping, and thus tend to predict the average of all possible target motions, easily resulting in plain/boring motions during inference. So we propose to explicitly model the one-to-many audio-to-motion mapping by splitting the cross-modal latent code into shared code and motion-specific code. The shared code is expected to be responsible for the motion component that is more correlated to the audio while the motion-specific code is expected to capture diverse motion information that is more independent of the audio. However, splitting the latent code into two parts poses extra training difficulties. Several crucial training losses/strategies, including relaxed motion loss, bicycle constraint, and diversity loss, are designed to better train the VAE. Experiments on both 3D and 2D motion datasets verify that our method generates more realistic and diverse motions than previous state-of-the-art methods, quantitatively and qualitatively. Besides, our formulation is compatible with discrete cosine transformation (DCT) modeling and other popular backbones (i.e., RNN, Transformer). As for motion losses and quantitative motion evaluation, we find structured losses/metrics (e.g. STFT) that consider temporal and/or spatial context complement the most commonly used point-wise losses (e.g. PCK), resulting in better motion dynamics and more nuanced motion details. Finally, we demonstrate that our method can be readily used to generate motion sequences with user-specified motion clips on the timeline.

A good motion retargeting cannot be reached without reasonable consideration of source-target differences on both the skeleton and shape geometry levels. In this work, we propose a novel Residual RETargeting network (R2ET) structure, which relies on two neural modification modules, to adjust the source motions to fit the target skeletons and shapes progressively. In particular, a skeleton-aware module is introduced to preserve the source motion semantics. A shape-aware module is designed to perceive the geometries of target characters to reduce interpenetration and contact-missing. Driven by our explored distance-based losses that explicitly model the motion semantics and geometry, these two modules can learn residual motion modifications on the source motion to generate plausible retargeted motion in a single inference without postprocessing. To balance these two modifications, we further present a balancing gate to conduct linear interpolation between them. Extensive experiments on the public dataset Mixamo demonstrate that our R2ET achieves the state-of-the-art performance, and provides a good balance between the preservation of motion semantics as well as the attenuation of interpenetration and contact-missing. Code is available at https://github.com/Kebii/R2ET.

We propose WarpingGAN, an effective and efficient 3D point cloud generation network. Unlike existing methods that generate point clouds by directly learning the mapping functions between latent codes and 3D shapes, Warping-GAN learns a unified local-warping function to warp multiple identical pre-defined priors (i.e., sets of points uniformly distributed on regular 3D grids) into 3D shapes driven by local structure-aware semantics. In addition, we also in-geniously utilize the principle of the discriminator and tai-lor a stitching loss to eliminate the gaps between different partitions of a generated shape corresponding to different priors for boosting quality. Owing to the novel gen-erating mechanism, WarpingGAN, a single lightweight network after one-time training, is capable of efficiently gen-erating uniformly distributed 3D point clouds with various resolutions. Extensive experimental results demonstrate the superiority of our WarpingGAN over state-of-the-art methods in terms of quantitative metrics, visual quality, and efficiency. The source code is publicly available at https://github.com/yztang4/WarpingGAN.git.

We present a fully automatic system that can produce high-fidelity, photo-realistic three-dimensional (3D) digital human heads with a consumer RGB-D selfie camera. The system only needs the user to take a short selfie RGB-D video while rotating his/her head and can produce a high-quality head reconstruction in less than 30 s. Our main contribution is a new facial geometry modeling and reflectance synthesis procedure that significantly improves the state of the art. Specifically, given the input video a two-stage frame selection procedure is first employed to select a few high-quality frames for reconstruction. Then a differentiable renderer-based 3D Morphable Model (3DMM) fitting algorithm is applied to recover facial geometries from multiview RGB-D data, which takes advantages of a powerful 3DMM basis constructed with extensive data generation and perturbation. Our 3DMM has much larger expressive capacities than conventional 3DMM, allowing us to recover more accurate facial geometry using merely linear basis. For reflectance synthesis, we present a hybrid approach that combines parametric fitting andConvolutional Neural Networks (CNNs)to synthesize high-resolution albedo/normal maps with realistic hair/pore/wrinkle details. Results show that our system can produce faithful 3D digital human faces with extremely realistic details. The main code and the newly constructed 3DMM basis is publicly available.

Reconstructing a 3D hand from a single-view RGB image is challenging due to various hand configurations and depth ambiguity. To reliably reconstruct a 3D hand from a monocular image, most state-of-the-art methods heavily rely on 3D annotations at the training stage, but obtaining 3D annotations is expensive. To alleviate reliance on labeled training data, we propose S2HAND, a self-supervised 3D hand reconstruction network that can jointly estimate pose, shape, texture, and the camera viewpoint. Specifically, we obtain geometric cues from the input image through easily accessible 2D detected keypoints. To learn an accurate hand reconstruction model from these noisy geometric cues, we utilize the consistency between 2D and 3D representations and propose a set of novel losses to rationalize outputs of the neural network. For the first time, we demonstrate the feasibility of training an accurate 3D hand reconstruction network without relying on manual annotations. Our experiments show that the proposed self-supervised method achieves comparable performance with recent fully-supervised methods. The code is available at https://github.com/TerenceCYJ/S2HAND.

Generating conversational gestures from speech audio is challenging due to the inherent one-to-many mapping be-tween audio and body motions. Conventional CNNs/RNNs assume one-to-one mapping, and thus tend to predict the average of all possible target motions, resulting in plain/boring motions during inference. In order to over-come this problem, we propose a novel conditional variational autoencoder (VAE) that explicitly models one-to-many audio-to-motion mapping by splitting the cross-modal latent code into shared code and motion-specific code. The shared code mainly models the strong correlation between audio and motion (such as the synchronized audio and motion beats), while the motion-specific code captures diverse motion information independent of the audio. However, splitting the latent code into two parts poses training difficulties for the VAE model. A mapping network facilitating random sampling along with other techniques including relaxed motion loss, bicycle constraint, and diversity loss are designed to better train the VAE. Experiments on both 3D and 2D motion datasets verify that our method generates more realistic and diverse motions than state-of-the-art methods, quantitatively and qualitatively. Finally, we demonstrate that our method can be readily used to generate motion sequences with user-specified motion clips on the timeline. Code and more results are at https://jingli513.github.io/audio2gestures.

Deep supervised hashing has emerged as an effective solution to large-scale semantic image retrieval problems in computer vision. Convolutional neural network-based hashing methods typically seek pairwise or triplet labels to conduct similarity-preserving learning. However, complex semantic concepts of visual contents are hard to capture by similar/dissimilar labels, which limits the retrieval performance. Generally, pairwise or triplet losses not only suffer from expensive training costs but also lack sufficient semantic information. In this paper, we propose a novel deep supervised hashing model to learn more compact class-level similarity-preserving binary codes. Our model is motivated by deep metric learning that directly takes semantic labels as supervised information in training and generates corresponding discriminant hashing code. Specifically, a novel cubic constraint loss function based on Gaussian distribution is proposed, which preserves semantic variations while penalizes the overlapping part of different classes in the embedding space. To address the discrete optimization problem introduced by binary codes, a two-step optimization strategy is proposed to provide efficient training and avoid the problem of gradient vanishing. Extensive experiments on five large-scale benchmark databases show that our model can achieve the state-of-the-art retrieval performance.

In this paper, we propose an effective training method to improve the performance of L2-normalized softmax for convolutional neural networks. Recent studies of deep learning show that by L2-normalizing the input features of softmax, the accuracy of CNN can be increased. Several works proposed novel loss functions based on the L2-normalized softmax. A common property shared by these modified normalized softmax models is that an extra set of parameters is introduced as the class centers. Although the physical meaning of this parameter is clear, few attentions have been paid to how to learn these class centers, which limits further improvement. In this paper, we address the problem of learning the class centers in the L2-normalized softmax. By treating the CNN training process as a time series, we propose a novel learning algorithm that combines the generally used gradient descent with the exponential moving average. Extensive experiments show that our model not only achieves better performance but also has a higher tolerance to the imbalance data.