Mengyuan Liu

Qitao Zhao, Ce Zheng, Mengyuan Liu et al.

Recently, transformer-based methods have gained significant success in sequential 2D-to-3D lifting human pose estimation. As a pioneering work, PoseFormer captures spatial relations of human joints in each video frame and human dynamics across frames with cascaded transformer layers and has achieved impressive performance. However, in real scenarios, the performance of PoseFormer and its follow-ups is limited by two factors: (a) The length of the input joint sequence; (b) The quality of 2D joint detection. Existing methods typically apply self-attention to all frames of the input sequence, causing a huge computational burden when the frame number is increased to obtain advanced estimation accuracy, and they are not robust to noise naturally brought by the limited capability of 2D joint detectors. In this paper, we propose PoseFormerV2, which exploits a compact representation of lengthy skeleton sequences in the frequency domain to efficiently scale up the receptive field and boost robustness to noisy 2D joint detection. With minimum modifications to PoseFormer, the proposed method effectively fuses features both in the time domain and frequency domain, enjoying a better speed-accuracy trade-off than its precursor. Extensive experiments on two benchmark datasets (i.e., Human3.6M and MPI-INF-3DHP) demonstrate that the proposed approach significantly outperforms the original PoseFormer and other transformer-based variants. Code is released at \url{https://github.com/QitaoZhao/PoseFormerV2}.

Lijun Li, Linrui Tian, Xindi Zhang et al.

The current interacting hand (IH) datasets are relatively simplistic in terms of background and texture, with hand joints being annotated by a machine annotator, which may result in inaccuracies, and the diversity of pose distribution is limited. However, the variability of background, pose distribution, and texture can greatly influence the generalization ability. Therefore, we present a large-scale synthetic dataset RenderIH for interacting hands with accurate and diverse pose annotations. The dataset contains 1M photo-realistic images with varied backgrounds, perspectives, and hand textures. To generate natural and diverse interacting poses, we propose a new pose optimization algorithm. Additionally, for better pose estimation accuracy, we introduce a transformer-based pose estimation network, TransHand, to leverage the correlation between interacting hands and verify the effectiveness of RenderIH in improving results. Our dataset is model-agnostic and can improve more accuracy of any hand pose estimation method in comparison to other real or synthetic datasets. Experiments have shown that pretraining on our synthetic data can significantly decrease the error from 6.76mm to 5.79mm, and our Transhand surpasses contemporary methods. Our dataset and code are available at https://github.com/adwardlee/RenderIH.

Wenhao Li, Mengyuan Liu, Hong Liu et al.

Modern multi-layer perceptron (MLP) models have shown competitive results in learning visual representations without self-attention. However, existing MLP models are not good at capturing local details and lack prior knowledge of human body configurations, which limits their modeling power for skeletal representation learning. To address these issues, we propose a simple yet effective graph-reinforced MLP-Like architecture, named GraphMLP, that combines MLPs and graph convolutional networks (GCNs) in a global-local-graphical unified architecture for 3D human pose estimation. GraphMLP incorporates the graph structure of human bodies into an MLP model to meet the domain-specific demand of the 3D human pose, while allowing for both local and global spatial interactions. Furthermore, we propose to flexibly and efficiently extend the GraphMLP to the video domain and show that complex temporal dynamics can be effectively modeled in a simple way with negligible computational cost gains in the sequence length. To the best of our knowledge, this is the first MLP-Like architecture for 3D human pose estimation in a single frame and a video sequence. Extensive experiments show that the proposed GraphMLP achieves state-of-the-art performance on two datasets, i.e., Human3.6M and MPI-INF-3DHP. Code and models are available at https://github.com/Vegetebird/GraphMLP.

Wenhao Li, Mengyuan Liu, Hong Liu et al.

Transformers have been successfully applied in the field of video-based 3D human pose estimation. However, the high computational costs of these video pose transformers (VPTs) make them impractical on resource-constrained devices. In this paper, we present a plug-and-play pruning-and-recovering framework, called Hourglass Tokenizer (HoT), for efficient transformer-based 3D human pose estimation from videos. Our HoT begins with pruning pose tokens of redundant frames and ends with recovering full-length tokens, resulting in a few pose tokens in the intermediate transformer blocks and thus improving the model efficiency. To effectively achieve this, we propose a token pruning cluster (TPC) that dynamically selects a few representative tokens with high semantic diversity while eliminating the redundancy of video frames. In addition, we develop a token recovering attention (TRA) to restore the detailed spatio-temporal information based on the selected tokens, thereby expanding the network output to the original full-length temporal resolution for fast inference. Extensive experiments on two benchmark datasets (i.e., Human3.6M and MPI-INF-3DHP) demonstrate that our method can achieve both high efficiency and estimation accuracy compared to the original VPT models. For instance, applying to MotionBERT and MixSTE on Human3.6M, our HoT can save nearly 50% FLOPs without sacrificing accuracy and nearly 40% FLOPs with only 0.2% accuracy drop, respectively. Code and models are available at https://github.com/NationalGAILab/HoT.

Yuhang Wen, Zixuan Tang, Yunsheng Pang et al.

Recognizing interactive action plays an important role in human-robot interaction and collaboration. Previous methods use late fusion and co-attention mechanism to capture interactive relations, which have limited learning capability or inefficiency to adapt to more interacting entities. With assumption that priors of each entity are already known, they also lack evaluations on a more general setting addressing the diversity of subjects. To address these problems, we propose an Interactive Spatiotemporal Token Attention Network (ISTA-Net), which simultaneously model spatial, temporal, and interactive relations. Specifically, our network contains a tokenizer to partition Interactive Spatiotemporal Tokens (ISTs), which is a unified way to represent motions of multiple diverse entities. By extending the entity dimension, ISTs provide better interactive representations. To jointly learn along three dimensions in ISTs, multi-head self-attention blocks integrated with 3D convolutions are designed to capture inter-token correlations. When modeling correlations, a strict entity ordering is usually irrelevant for recognizing interactive actions. To this end, Entity Rearrangement is proposed to eliminate the orderliness in ISTs for interchangeable entities. Extensive experiments on four datasets verify the effectiveness of ISTA-Net by outperforming state-of-the-art methods. Our code is publicly available at https://github.com/Necolizer/ISTA-Net

Bin Ren, Guofeng Mei, Danda Pani Paudel et al.

Contrastive learning (CL) for Vision Transformers (ViTs) in image domains has achieved performance comparable to CL for traditional convolutional backbones. However, in 3D point cloud pretraining with ViTs, masked autoencoder (MAE) modeling remains dominant. This raises the question: Can we take the best of both worlds? To answer this question, we first empirically validate that integrating MAE-based point cloud pre-training with the standard contrastive learning paradigm, even with meticulous design, can lead to a decrease in performance. To address this limitation, we reintroduce CL into the MAE-based point cloud pre-training paradigm by leveraging the inherent contrastive properties of MAE. Specifically, rather than relying on extensive data augmentation as commonly used in the image domain, we randomly mask the input tokens twice to generate contrastive input pairs. Subsequently, a weight-sharing encoder and two identically structured decoders are utilized to perform masked token reconstruction. Additionally, we propose that for an input token masked by both masks simultaneously, the reconstructed features should be as similar as possible. This naturally establishes an explicit contrastive constraint within the generative MAE-based pre-training paradigm, resulting in our proposed method, Point-CMAE. Consequently, Point-CMAE effectively enhances the representation quality and transfer performance compared to its MAE counterpart. Experimental evaluations across various downstream applications, including classification, part segmentation, and few-shot learning, demonstrate the efficacy of our framework in surpassing state-of-the-art techniques under standard ViTs and single-modal settings. The source code and trained models are available at: https://github.com/Amazingren/Point-CMAE.

Hongbo Kang, Yong Wang, Mengyuan Liu et al.

Reconstructing 3D poses from 2D poses lacking depth information is particularly challenging due to the complexity and diversity of human motion. The key is to effectively model the spatial constraints between joints to leverage their inherent dependencies. Thus, we propose a novel model, called Double-chain Graph Convolutional Transformer (DC-GCT), to constrain the pose through a double-chain design consisting of local-to-global and global-to-local chains to obtain a complex representation more suitable for the current human pose. Specifically, we combine the advantages of GCN and Transformer and design a Local Constraint Module (LCM) based on GCN and a Global Constraint Module (GCM) based on self-attention mechanism as well as a Feature Interaction Module (FIM). The proposed method fully captures the multi-level dependencies between human body joints to optimize the modeling capability of the model. Moreover, we propose a method to use temporal information into the single-frame model by guiding the video sequence embedding through the joint embedding of the target frame, with negligible increase in computational cost. Experimental results demonstrate that DC-GCT achieves state-of-the-art performance on two challenging datasets (Human3.6M and MPI-INF-3DHP). Notably, our model achieves state-of-the-art performance on all action categories in the Human3.6M dataset using detected 2D poses from CPN, and our code is available at: https://github.com/KHB1698/DC-GCT.

Runwei Ding, Yuhang Wen, Jinfu Liu et al.

Human skeletons and RGB sequences are both widely-adopted input modalities for human action recognition. However, skeletons lack appearance features and color data suffer large amount of irrelevant depiction. To address this, we introduce human parsing feature map as a novel modality, since it can selectively retain spatiotemporal features of the body parts, while filtering out noises regarding outfits, backgrounds, etc. We propose an Integrating Human Parsing and Pose Network (IPP-Net) for action recognition, which is the first to leverage both skeletons and human parsing feature maps in dual-branch approach. The human pose branch feeds compact skeletal representations of different modalities in graph convolutional network to model pose features. In human parsing branch, multi-frame body-part parsing features are extracted with human detector and parser, which is later learnt using a convolutional backbone. A late ensemble of two branches is adopted to get final predictions, considering both robust keypoints and rich semantic body-part features. Extensive experiments on NTU RGB+D and NTU RGB+D 120 benchmarks consistently verify the effectiveness of the proposed IPP-Net, which outperforms the existing action recognition methods. Our code is publicly available at https://github.com/liujf69/IPP-Net-Parsing .

Zhongbin Fang, Xiangtai Li, Xia Li et al.

With the rise of large-scale models trained on broad data, in-context learning has become a new learning paradigm that has demonstrated significant potential in natural language processing and computer vision tasks. Meanwhile, in-context learning is still largely unexplored in the 3D point cloud domain. Although masked modeling has been successfully applied for in-context learning in 2D vision, directly extending it to 3D point clouds remains a formidable challenge. In the case of point clouds, the tokens themselves are the point cloud positions (coordinates) that are masked during inference. Moreover, position embedding in previous works may inadvertently introduce information leakage. To address these challenges, we introduce a novel framework, named Point-In-Context, designed especially for in-context learning in 3D point clouds, where both inputs and outputs are modeled as coordinates for each task. Additionally, we propose the Joint Sampling module, carefully designed to work in tandem with the general point sampling operator, effectively resolving the aforementioned technical issues. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks.

Peiming Li, Ziyi Wang, Mengyuan Liu et al.

Grasp generation aims to create complex hand-object interactions with a specified object. While traditional approaches for hand generation have primarily focused on visibility and diversity under scene constraints, they tend to overlook the fine-grained hand-object interactions such as contacts, resulting in inaccurate and undesired grasps. To address these challenges, we propose a controllable grasp generation task and introduce ClickDiff, a controllable conditional generation model that leverages a fine-grained Semantic Contact Map (SCM). Particularly when synthesizing interactive grasps, the method enables the precise control of grasp synthesis through either user-specified or algorithmically predicted Semantic Contact Map. Specifically, to optimally utilize contact supervision constraints and to accurately model the complex physical structure of hands, we propose a Dual Generation Framework. Within this framework, the Semantic Conditional Module generates reasonable contact maps based on fine-grained contact information, while the Contact Conditional Module utilizes contact maps alongside object point clouds to generate realistic grasps. We evaluate the evaluation criteria applicable to controllable grasp generation. Both unimanual and bimanual generation experiments on GRAB and ARCTIC datasets verify the validity of our proposed method, demonstrating the efficacy and robustness of ClickDiff, even with previously unseen objects. Our code is available at https://github.com/adventurer-w/ClickDiff.

Jiaqi Wang, Hanqi Jiang, Yiheng Liu et al.

In an era defined by the explosive growth of data and rapid technological advancements, Multimodal Large Language Models (MLLMs) stand at the forefront of artificial intelligence (AI) systems. Designed to seamlessly integrate diverse data types-including text, images, videos, audio, and physiological sequences-MLLMs address the complexities of real-world applications far beyond the capabilities of single-modality systems. In this paper, we systematically sort out the applications of MLLM in multimodal tasks such as natural language, vision, and audio. We also provide a comparative analysis of the focus of different MLLMs in the tasks, and provide insights into the shortcomings of current MLLMs, and suggest potential directions for future research. Through these discussions, this paper hopes to provide valuable insights for the further development and application of MLLM.

Tao Wang, Mengyuan Liu, Hong Liu et al.

Occluded person re-identification (Re-ID) is a challenging problem due to the destruction of occluders. Most existing methods focus on visible human body parts through some prior information. However, when complementary occlusions occur, features in occluded regions can interfere with matching, which affects performance severely. In this paper, different from most previous works that discard the occluded region, we propose a Feature Completion Transformer (FCFormer) to implicitly complement the semantic information of occluded parts in the feature space. Specifically, Occlusion Instance Augmentation (OIA) is proposed to simulates real and diverse occlusion situations on the holistic image. These augmented images not only enrich the amount of occlusion samples in the training set, but also form pairs with the holistic images. Subsequently, a dual-stream architecture with a shared encoder is proposed to learn paired discriminative features from pairs of inputs. Without additional semantic information, an occluded-holistic feature sample-label pair can be automatically created. Then, Feature Completion Decoder (FCD) is designed to complement the features of occluded regions by using learnable tokens to aggregate possible information from self-generated occluded features. Finally, we propose the Cross Hard Triplet (CHT) loss to further bridge the gap between complementing features and extracting features under the same ID. In addition, Feature Completion Consistency (FC$^2$) loss is introduced to help the generated completion feature distribution to be closer to the real holistic feature distribution. Extensive experiments over five challenging datasets demonstrate that the proposed FCFormer achieves superior performance and outperforms the state-of-the-art methods by significant margins on occluded datasets.

Yi Zhang, Youjun Zhao, Yuhang Wen et al.

Spotting facial micro-expression from videos finds various potential applications in fields including clinical diagnosis and interrogation, meanwhile this task is still difficult due to the limited scale of training data. To solve this problem, this paper tries to formulate a new task called micro-expression generation and then presents a strong baseline which combines the first order motion model with facial prior knowledge. Given a target face, we intend to drive the face to generate micro-expression videos according to the motion patterns of source videos. Specifically, our new model involves three modules. First, we extract facial prior features from a region focusing module. Second, we estimate facial motion using key points and local affine transformations with a motion prediction module. Third, expression generation module is used to drive the target face to generate videos. We train our model on public CASME II, SAMM and SMIC datasets and then use the model to generate new micro-expression videos for evaluation. Our model achieves the first place in the Facial Micro-Expression Challenge 2021 (MEGC2021), where our superior performance is verified by three experts with Facial Action Coding System certification. Source code is provided in https://github.com/Necolizer/Facial-Prior-Based-FOMM.

Jinfu Liu, Chen Chen, Mengyuan Liu

Skeleton-based action recognition has garnered significant attention due to the utilization of concise and resilient skeletons. Nevertheless, the absence of detailed body information in skeletons restricts performance, while other multimodal methods require substantial inference resources and are inefficient when using multimodal data during both training and inference stages. To address this and fully harness the complementary multimodal features, we propose a novel multi-modality co-learning (MMCL) framework by leveraging the multimodal large language models (LLMs) as auxiliary networks for efficient skeleton-based action recognition, which engages in multi-modality co-learning during the training stage and keeps efficiency by employing only concise skeletons in inference. Our MMCL framework primarily consists of two modules. First, the Feature Alignment Module (FAM) extracts rich RGB features from video frames and aligns them with global skeleton features via contrastive learning. Second, the Feature Refinement Module (FRM) uses RGB images with temporal information and text instruction to generate instructive features based on the powerful generalization of multimodal LLMs. These instructive text features will further refine the classification scores and the refined scores will enhance the model's robustness and generalization in a manner similar to soft labels. Extensive experiments on NTU RGB+D, NTU RGB+D 120 and Northwestern-UCLA benchmarks consistently verify the effectiveness of our MMCL, which outperforms the existing skeleton-based action recognition methods. Meanwhile, experiments on UTD-MHAD and SYSU-Action datasets demonstrate the commendable generalization of our MMCL in zero-shot and domain-adaptive action recognition. Our code is publicly available at: https://github.com/liujf69/MMCL-Action.

Linhui Dai, Hong Liu, Pinhao Song et al.

Underwater object detection (UOD) plays a significant role in aquaculture and marine environmental protection. Considering the challenges posed by low contrast and low-light conditions in underwater environments, several underwater image enhancement (UIE) methods have been proposed to improve the quality of underwater images. However, only using the enhanced images does not improve the performance of UOD, since it may unavoidably remove or alter critical patterns and details of underwater objects. In contrast, we believe that exploring the complementary information from the two domains is beneficial for UOD. The raw image preserves the natural characteristics of the scene and texture information of the objects, while the enhanced image improves the visibility of underwater objects. Based on this perspective, we propose a Gated Cross-domain Collaborative Network (GCC-Net) to address the challenges of poor visibility and low contrast in underwater environments, which comprises three dedicated components. Firstly, a real-time UIE method is employed to generate enhanced images, which can improve the visibility of objects in low-contrast areas. Secondly, a cross-domain feature interaction module is introduced to facilitate the interaction and mine complementary information between raw and enhanced image features. Thirdly, to prevent the contamination of unreliable generated results, a gated feature fusion module is proposed to adaptively control the fusion ratio of cross-domain information. Our method presents a new UOD paradigm from the perspective of cross-domain information interaction and fusion. Experimental results demonstrate that the proposed GCC-Net achieves state-of-the-art performance on four underwater datasets.

Mengyuan Liu, Ziyi Wang, Peiming Li et al.

RGB camera-based surveillance systems enable human action recognition for public safety and healthcare, yet raise serious privacy concerns. Existing methods rely on post-capture algorithms, which fail to protect privacy during data acquisition. We propose Lens Privacy Sealing (LPS), a simple hardware solution that physically obscures camera lenses with adjustable laminating film, providing pre-sensor privacy protection at minimal cost. Unlike software methods or expensive engineered optics, LPS achieves strong privacy through stochastic multi-layer scattering that is physically irreversible. We introduce the P$^3$AR dataset for privacy-preserving action recognition, featuring both large-scale replay-captured (P$^3$AR-NTU, 114K videos) and real-world collected (P$^3$AR-PKU) subsets with privacy attribute annotations. To handle video degradation from LPS, we propose MSPNet, a single-stage framework incorporating Inter-Frame Noise Suppressor (IFNS) and Cross-Frame Semantic Aggregator (CFSA), enhanced by contrastive language-image pre-training for robust semantic extraction. Extensive experiments demonstrate that MSPNet with IFNS and CFSA nearly doubles action recognition accuracy compared to baseline methods while suppressing identity recognition to low levels. Comprehensive validation shows LPS achieves a superior privacy-utility trade-off compared to state-of-the-art hardware methods, resists reconstruction attacks including PSF inversion and data-driven recovery, and generalizes robustly across optical configurations and challenging environments. Code is available at https://github.com/wangzy01/MSPNet.

Yang Liu, Chen Chen, Can Wang et al.

Masked Autoencoders (MAE) have demonstrated promising performance in self-supervised learning for both 2D and 3D computer vision. Nevertheless, existing MAE-based methods still have certain drawbacks. Firstly, the functional decoupling between the encoder and decoder is incomplete, which limits the encoder's representation learning ability. Secondly, downstream tasks solely utilize the encoder, failing to fully leverage the knowledge acquired through the encoder-decoder architecture in the pre-text task. In this paper, we propose Point Regress AutoEncoder (Point-RAE), a new scheme for regressive autoencoders for point cloud self-supervised learning. The proposed method decouples functions between the decoder and the encoder by introducing a mask regressor, which predicts the masked patch representation from the visible patch representation encoded by the encoder and the decoder reconstructs the target from the predicted masked patch representation. By doing so, we minimize the impact of decoder updates on the representation space of the encoder. Moreover, we introduce an alignment constraint to ensure that the representations for masked patches, predicted from the encoded representations of visible patches, are aligned with the masked patch presentations computed from the encoder. To make full use of the knowledge learned in the pre-training stage, we design a new finetune mode for the proposed Point-RAE. Extensive experiments demonstrate that our approach is efficient during pre-training and generalizes well on various downstream tasks. Specifically, our pre-trained models achieve a high accuracy of \textbf{90.28\%} on the ScanObjectNN hardest split and \textbf{94.1\%} accuracy on ModelNet40, surpassing all the other self-supervised learning methods. Our code and pretrained model are public available at: \url{https://github.com/liuyyy111/Point-RAE}.

Mengyi Zhao, Mengyuan Liu, Bin Ren et al.

Diffusion-based generative models have recently emerged as powerful solutions for high-quality synthesis in multiple domains. Leveraging the bidirectional Markov chains, diffusion probabilistic models generate samples by inferring the reversed Markov chain based on the learned distribution mapping at the forward diffusion process. In this work, we propose Modiff, a conditional paradigm that benefits from the denoising diffusion probabilistic model (DDPM) to tackle the problem of realistic and diverse action-conditioned 3D skeleton-based motion generation. We are a pioneering attempt that uses DDPM to synthesize a variable number of motion sequences conditioned on a categorical action. We evaluate our approach on the large-scale NTU RGB+D dataset and show improvements over state-of-the-art motion generation methods.

Xinshun Wang, Qiongjie Cui, Chen Chen et al.

Existing Graph Convolutional Networks to achieve human motion prediction largely adopt a one-step scheme, which output the prediction straight from history input, failing to exploit human motion patterns. We observe that human motions have transitional patterns and can be split into snippets representative of each transition. Each snippet can be reconstructed from its starting and ending poses referred to as the transitional poses. We propose a snippet-to-motion multi-stage framework that breaks motion prediction into sub-tasks easier to accomplish. Each sub-task integrates three modules: transitional pose prediction, snippet reconstruction, and snippet-to-motion prediction. Specifically, we propose to first predict only the transitional poses. Then we use them to reconstruct the corresponding snippets, obtaining a close approximation to the true motion sequence. Finally we refine them to produce the final prediction output. To implement the network, we propose a novel unified graph modeling, which allows for direct and effective feature propagation compared to existing approaches which rely on separate space-time modeling. Extensive experiments on Human 3.6M, CMU Mocap and 3DPW datasets verify the effectiveness of our method which achieves state-of-the-art performance.

Mengyuan Liu, Jinfu Liu, Yongkang Jiang et al.

Human action recognition (HAR) in videos has garnered widespread attention due to the rich information in RGB videos. Nevertheless, existing methods for extracting deep features from RGB videos face challenges such as information redundancy, susceptibility to noise and high storage costs. To address these issues and fully harness the useful information in videos, we propose a novel heatmap pooling network (HP-Net) for action recognition from videos, which extracts information-rich, robust and concise pooled features of the human body in videos through a feedback pooling module. The extracted pooled features demonstrate obvious performance advantages over the previously obtained pose data and heatmap features from videos. In addition, we design a spatial-motion co-learning module and a text refinement modulation module to integrate the extracted pooled features with other multimodal data, enabling more robust action recognition. Extensive experiments on several benchmarks namely NTU RGB+D 60, NTU RGB+D 120, Toyota-Smarthome and UAV-Human consistently verify the effectiveness of our HP-Net, which outperforms the existing human action recognition methods. Our code is publicly available at: https://github.com/liujf69/HPNet-Action.

Renliang Sun, Mengyuan Liu, Shiping Yang et al.

Benefiting from diverse instruction datasets, contemporary Large Language Models (LLMs) perform effectively as AI assistants in collaborating with humans. However, LLMs still struggle to generate natural and colloquial responses in real-world applications such as chatbots and psychological counseling that require more human-like interactions. To address these limitations, we introduce NICO, a Natural Interactive COnversation dataset in Chinese. We first use GPT-4-turbo to generate dialogue drafts and make them cover 20 daily-life topics and 5 types of social interactions. Then, we hire workers to revise these dialogues to ensure that they are free of grammatical errors and unnatural utterances. We define two dialogue-level natural conversation tasks and two sentence-level tasks for identifying and rewriting unnatural sentences. Multiple open-source and closed-source LLMs are tested and analyzed in detail. The experimental results highlight the challenge of the tasks and demonstrate how NICO can help foster the natural dialogue capabilities of LLMs. The dataset will be released.

Mengyuan Liu, Hong Liu, Tianyu Guo

Considering the instance-level discriminative ability, contrastive learning methods, including MoCo and SimCLR, have been adapted from the original image representation learning task to solve the self-supervised skeleton-based action recognition task. These methods usually use multiple data streams (i.e., joint, motion, and bone) for ensemble learning, meanwhile, how to construct a discriminative feature space within a single stream and effectively aggregate the information from multiple streams remains an open problem. To this end, this paper first applies a new contrastive learning method called BYOL to learn from skeleton data, and then formulate SkeletonBYOL as a simple yet effective baseline for self-supervised skeleton-based action recognition. Inspired by SkeletonBYOL, this paper further presents a Cross-Model and Cross-Stream (CMCS) framework. This framework combines Cross-Model Adversarial Learning (CMAL) and Cross-Stream Collaborative Learning (CSCL). Specifically, CMAL learns single-stream representation by cross-model adversarial loss to obtain more discriminative features. To aggregate and interact with multi-stream information, CSCL is designed by generating similarity pseudo label of ensemble learning as supervision and guiding feature generation for individual streams. Extensive experiments on three datasets verify the complementary properties between CMAL and CSCL and also verify that the proposed method can achieve better results than state-of-the-art methods using various evaluation protocols.

Qitao Zhao, Ce Zheng, Mengyuan Liu et al.

The dominant paradigm in 3D human pose estimation that lifts a 2D pose sequence to 3D heavily relies on long-term temporal clues (i.e., using a daunting number of video frames) for improved accuracy, which incurs performance saturation, intractable computation and the non-causal problem. This can be attributed to their inherent inability to perceive spatial context as plain 2D joint coordinates carry no visual cues. To address this issue, we propose a straightforward yet powerful solution: leveraging the readily available intermediate visual representations produced by off-the-shelf (pre-trained) 2D pose detectors -- no finetuning on the 3D task is even needed. The key observation is that, while the pose detector learns to localize 2D joints, such representations (e.g., feature maps) implicitly encode the joint-centric spatial context thanks to the regional operations in backbone networks. We design a simple baseline named Context-Aware PoseFormer to showcase its effectiveness. Without access to any temporal information, the proposed method significantly outperforms its context-agnostic counterpart, PoseFormer, and other state-of-the-art methods using up to hundreds of video frames regarding both speed and precision. Project page: https://qitaozhao.github.io/ContextAware-PoseFormer

Xinshun Wang, Qiongjie Cui, Chen Chen et al.

In recent years, Graph Convolutional Networks (GCNs) have been widely used in human motion prediction, but their performance remains unsatisfactory. Recently, MLP-Mixer, initially developed for vision tasks, has been leveraged into human motion prediction as a promising alternative to GCNs, which achieves both better performance and better efficiency than GCNs. Unlike GCNs, which can explicitly capture human skeleton's bone-joint structure by representing it as a graph with edges and nodes, MLP-Mixer relies on fully connected layers and thus cannot explicitly model such graph-like structure of human's. To break this limitation of MLP-Mixer's, we propose \textit{Graph-Guided Mixer}, a novel approach that equips the original MLP-Mixer architecture with the capability to model graph structure. By incorporating graph guidance, our \textit{Graph-Guided Mixer} can effectively capture and utilize the specific connectivity patterns within human skeleton's graph representation. In this paper, first we uncover a theoretical connection between MLP-Mixer and GCN that is unexplored in existing research. Building on this theoretical connection, next we present our proposed \textit{Graph-Guided Mixer}, explaining how the original MLP-Mixer architecture is reinvented to incorporate guidance from graph structure. Then we conduct an extensive evaluation on the Human3.6M, AMASS, and 3DPW datasets, which shows that our method achieves state-of-the-art performance.

Wanying Zhang, Shen Zhao, Fanyang Meng et al.

With potential applications in fields including intelligent surveillance and human-robot interaction, the human motion prediction task has become a hot research topic and also has achieved high success, especially using the recent Graph Convolutional Network (GCN). Current human motion prediction task usually focuses on predicting human motions for atomic actions. Observing that atomic actions can happen at the same time and thus formulating the composite actions, we propose the composite human motion prediction task. To handle this task, we first present a Composite Action Generation (CAG) module to generate synthetic composite actions for training, thus avoiding the laborious work of collecting composite action samples. Moreover, we alleviate the effect of composite actions on demand for a more complicated model by presenting a Dynamic Compositional Graph Convolutional Network (DC-GCN). Extensive experiments on the Human3.6M dataset and our newly collected CHAMP dataset consistently verify the efficiency of our DC-GCN method, which achieves state-of-the-art motion prediction accuracies and meanwhile needs few extra computational costs than traditional GCN-based human motion methods.

Xinshun Wang, Wanying Zhang, Can Wang et al.

Graph Convolutional Networks (GCN) which typically follows a neural message passing framework to model dependencies among skeletal joints has achieved high success in skeleton-based human motion prediction task. Nevertheless, how to construct a graph from a skeleton sequence and how to perform message passing on the graph are still open problems, which severely affect the performance of GCN. To solve both problems, this paper presents a Dynamic Dense Graph Convolutional Network (DD-GCN), which constructs a dense graph and implements an integrated dynamic message passing. More specifically, we construct a dense graph with 4D adjacency modeling as a comprehensive representation of motion sequence at different levels of abstraction. Based on the dense graph, we propose a dynamic message passing framework that learns dynamically from data to generate distinctive messages reflecting sample-specific relevance among nodes in the graph. Extensive experiments on benchmark Human 3.6M and CMU Mocap datasets verify the effectiveness of our DD-GCN which obviously outperforms state-of-the-art GCN-based methods, especially when using long-term and our proposed extremely long-term protocol.

Mengyuan Liu, Sheng Yan, Yong Wang et al.

We introduce MoSa, a novel hierarchical motion generation framework for text-driven 3D human motion generation that enhances the Vector Quantization-guided Generative Transformers (VQ-GT) paradigm through a coarse-to-fine scalable generation process. In MoSa, we propose a Multi-scale Token Preservation Strategy (MTPS) integrated into a hierarchical residual vector quantization variational autoencoder (RQ-VAE). MTPS employs interpolation at each hierarchical quantization to effectively retain coarse-to-fine multi-scale tokens. With this, the generative transformer supports Scalable Autoregressive (SAR) modeling, which predicts scale tokens, unlike traditional methods that predict only one token at each step. Consequently, MoSa requires only 10 inference steps, matching the number of RQ-VAE quantization layers. To address potential reconstruction degradation from frequent interpolation, we propose CAQ-VAE, a lightweight yet expressive convolution-attention hybrid VQ-VAE. CAQ-VAE enhances residual block design and incorporates attention mechanisms to better capture global dependencies. Extensive experiments show that MoSa achieves state-of-the-art generation quality and efficiency, outperforming prior methods in both fidelity and speed. On the Motion-X dataset, MoSa achieves an FID of 0.06 (versus MoMask's 0.20) while reducing inference time by 27 percent. Moreover, MoSa generalizes well to downstream tasks such as motion editing, requiring no additional fine-tuning. The code is available at https://mosa-web.github.io/MoSa-web

Jinfu Liu, Runwei Ding, Yuhang Wen et al.

Multimodal-based action recognition methods have achieved high success using pose and RGB modality. However, skeletons sequences lack appearance depiction and RGB images suffer irrelevant noise due to modality limitations. To address this, we introduce human parsing feature map as a novel modality, since it can selectively retain effective semantic features of the body parts, while filtering out most irrelevant noise. We propose a new dual-branch framework called Ensemble Human Parsing and Pose Network (EPP-Net), which is the first to leverage both skeletons and human parsing modalities for action recognition. The first human pose branch feeds robust skeletons in graph convolutional network to model pose features, while the second human parsing branch also leverages depictive parsing feature maps to model parsing festures via convolutional backbones. The two high-level features will be effectively combined through a late fusion strategy for better action recognition. Extensive experiments on NTU RGB+D and NTU RGB+D 120 benchmarks consistently verify the effectiveness of our proposed EPP-Net, which outperforms the existing action recognition methods. Our code is available at: https://github.com/liujf69/EPP-Net-Action.

You Wu, Xucheng Wang, Xiangyang Yang et al.

Single-stream architectures using Vision Transformer (ViT) backbones show great potential for real-time UAV tracking recently. However, frequent occlusions from obstacles like buildings and trees expose a major drawback: these models often lack strategies to handle occlusions effectively. New methods are needed to enhance the occlusion resilience of single-stream ViT models in aerial tracking. In this work, we propose to learn Occlusion-Robust Representations (ORR) based on ViTs for UAV tracking by enforcing an invariance of the feature representation of a target with respect to random masking operations modeled by a spatial Cox process. Hopefully, this random masking approximately simulates target occlusions, thereby enabling us to learn ViTs that are robust to target occlusion for UAV tracking. This framework is termed ORTrack. Additionally, to facilitate real-time applications, we propose an Adaptive Feature-Based Knowledge Distillation (AFKD) method to create a more compact tracker, which adaptively mimics the behavior of the teacher model ORTrack according to the task's difficulty. This student model, dubbed ORTrack-D, retains much of ORTrack's performance while offering higher efficiency. Extensive experiments on multiple benchmarks validate the effectiveness of our method, demonstrating its state-of-the-art performance. Codes is available at https://github.com/wuyou3474/ORTrack.

Zhichao Deng, Xiangtai Li, Xia Li et al.

Understanding the real world through point cloud video is a crucial aspect of robotics and autonomous driving systems. However, prevailing methods for 4D point cloud recognition have limitations due to sensor resolution, which leads to a lack of detailed information. Recent advances have shown that Vision-Language Models (VLM) pre-trained on web-scale text-image datasets can learn fine-grained visual concepts that can be transferred to various downstream tasks. However, effectively integrating VLM into the domain of 4D point clouds remains an unresolved problem. In this work, we propose the Vision-Language Models Goes 4D (VG4D) framework to transfer VLM knowledge from visual-text pre-trained models to a 4D point cloud network. Our approach involves aligning the 4D encoder's representation with a VLM to learn a shared visual and text space from training on large-scale image-text pairs. By transferring the knowledge of the VLM to the 4D encoder and combining the VLM, our VG4D achieves improved recognition performance. To enhance the 4D encoder, we modernize the classic dynamic point cloud backbone and propose an improved version of PSTNet, im-PSTNet, which can efficiently model point cloud videos. Experiments demonstrate that our method achieves state-of-the-art performance for action recognition on both the NTU RGB+D 60 dataset and the NTU RGB+D 120 dataset. Code is available at \url{https://github.com/Shark0-0/VG4D}.

Pengxiang Ding, Qiongjie Cui, Min Zhang et al.

Human motion forecasting, with the goal of estimating future human behavior over a period of time, is a fundamental task in many real-world applications. However, existing works typically concentrate on predicting the major joints of the human body without considering the delicate movements of the human hands. In practical applications, hand gesture plays an important role in human communication with the real world, and expresses the primary intention of human beings. In this work, we are the first to formulate a whole-body human pose forecasting task, which jointly predicts the future body and hand activities. Correspondingly, we propose a novel Encoding-Alignment-Interaction (EAI) framework that aims to predict both coarse (body joints) and fine-grained (gestures) activities collaboratively, enabling expressive and cross-facilitated forecasting of 3D whole-body human motions. Specifically, our model involves two key constituents: cross-context alignment (XCA) and cross-context interaction (XCI). Considering the heterogeneous information within the whole-body, XCA aims to align the latent features of various human components, while XCI focuses on effectively capturing the context interaction among the human components. We conduct extensive experiments on a newly-introduced large-scale benchmark and achieve state-of-the-art performance. The code is public for research purposes at https://github.com/Dingpx/EAI.

Jinfu Liu, Baiqiao Yin, Jiaying Lin et al.

Skeleton-based action recognition has gained considerable traction thanks to its utilization of succinct and robust skeletal representations. Nonetheless, current methodologies often lean towards utilizing a solitary backbone to model skeleton modality, which can be limited by inherent flaws in the network backbone. To address this and fully leverage the complementary characteristics of various network architectures, we propose a novel Hybrid Dual-Branch Network (HDBN) for robust skeleton-based action recognition, which benefits from the graph convolutional network's proficiency in handling graph-structured data and the powerful modeling capabilities of Transformers for global information. In detail, our proposed HDBN is divided into two trunk branches: MixGCN and MixFormer. The two branches utilize GCNs and Transformers to model both 2D and 3D skeletal modalities respectively. Our proposed HDBN emerged as one of the top solutions in the Multi-Modal Video Reasoning and Analyzing Competition (MMVRAC) of 2024 ICME Grand Challenge, achieving accuracies of 47.95% and 75.36% on two benchmarks of the UAV-Human dataset by outperforming most existing methods. Our code will be publicly available at: https://github.com/liujf69/ICMEW2024-Track10.

Hongbo Kang, Yong Wang, Mengyuan Liu et al.

Previous probabilistic models for 3D Human Pose Estimation (3DHPE) aimed to enhance pose accuracy by generating multiple hypotheses. However, most of the hypotheses generated deviate substantially from the true pose. Compared to deterministic models, the excessive uncertainty in probabilistic models leads to weaker performance in single-hypothesis prediction. To address these two challenges, we propose a diffusion-based refinement framework called DRPose, which refines the output of deterministic models by reverse diffusion and achieves more suitable multi-hypothesis prediction for the current pose benchmark by multi-step refinement with multiple noises. To this end, we propose a Scalable Graph Convolution Transformer (SGCT) and a Pose Refinement Module (PRM) for denoising and refining. Extensive experiments on Human3.6M and MPI-INF-3DHP datasets demonstrate that our method achieves state-of-the-art performance on both single and multi-hypothesis 3DHPE. Code is available at https://github.com/KHB1698/DRPose.

Sheng Yan, Mengyuan Liu, Yong Wang et al.

In this paper, we address the unexplored question of temporal sentence localization in human motions (TSLM), aiming to locate a target moment from a 3D human motion that semantically corresponds to a text query. Considering that 3D human motions are captured using specialized motion capture devices, motions with only a few joints lack complex scene information like objects and lighting. Due to this character, motion data has low contextual richness and semantic ambiguity between frames, which limits the accuracy of predictions made by current video localization frameworks extended to TSLM to only a rough level. To refine this, we devise two novel label-prior-assisted training schemes: one embed prior knowledge of foreground and background to highlight the localization chances of target moments, and the other forces the originally rough predictions to overlap with the more accurate predictions obtained from the flipped start/end prior label sequences during recovery training. We show that injecting label-prior knowledge into the model is crucial for improving performance at high IoU. In our constructed TSLM benchmark, our model termed MLP achieves a recall of 44.13 at IoU@0.7 on the BABEL dataset and 71.17 on HumanML3D (Restore), outperforming prior works. Finally, we showcase the potential of our approach in corpus-level moment retrieval. Our source code is openly accessible at https://github.com/eanson023/mlp.

Ziyi Wang, Xinshun Wang, Shuang Chen et al.

We present UniMotion, to our knowledge the first unified framework for simultaneous understanding and generation of human motion, natural language, and RGB images within a single architecture. Existing unified models handle only restricted modality subsets (e.g., Motion-Text or static Pose-Image) and predominantly rely on discrete tokenization, which introduces quantization errors and disrupts temporal continuity. UniMotion overcomes both limitations through a core principle: treating motion as a first-class continuous modality on equal footing with RGB. A novel Cross-Modal Aligned Motion VAE (CMA-VAE) and symmetric dual-path embedders construct parallel continuous pathways for Motion and RGB within a shared LLM backbone. To inject visual-semantic priors into motion representations without requiring images at inference, we propose Dual-Posterior KL Alignment (DPA), which distills a vision-fused encoder's richer posterior into the motion-only encoder. To address the cold-start problem -- where text supervision alone is too sparse to calibrate the newly introduced motion pathway -- we further propose Latent Reconstruction Alignment (LRA), a self-supervised pre-training strategy that uses dense motion latents as unambiguous conditions to co-calibrate the embedder, backbone, and flow head, establishing a stable motion-aware foundation for all downstream tasks. UniMotion achieves state-of-the-art performance across seven tasks spanning any-to-any understanding, generation, and editing among the three modalities, with especially strong advantages on cross-modal compositional tasks.

Yue Li, Baiqiao Yin, Jinfu Liu et al.

In recent years, Event Sound Source Localization has been widely applied in various fields. Recent works typically relying on the contrastive learning framework show impressive performance. However, all work is based on large relatively simple datasets. It's also crucial to understand and analyze human behaviors (actions and interactions of people), voices, and sounds in chaotic events in many applications, e.g., crowd management, and emergency response services. In this paper, we apply the existing model to a more complex dataset, explore the influence of parameters on the model, and propose a semi-supervised improvement method SemiPL. With the increase in data quantity and the influence of label quality, self-supervised learning will be an unstoppable trend. The experiment shows that the parameter adjustment will positively affect the existing model. In particular, SSPL achieved an improvement of 12.2% cIoU and 0.56% AUC in Chaotic World compared to the results provided. The code is available at: https://github.com/ly245422/SSPL

Jiaying Lin, Jiajun Wen, Mengyuan Liu et al.

The task of spatiotemporal action localization in chaotic scenes is a challenging task toward advanced video understanding. Paving the way with high-quality video feature extraction and enhancing the precision of detector-predicted anchors can effectively improve model performance. To this end, we propose a high-performance dual-stream spatiotemporal feature extraction network SFMViT with an anchor pruning strategy. The backbone of our SFMViT is composed of ViT and SlowFast with prior knowledge of spatiotemporal action localization, which fully utilizes ViT's excellent global feature extraction capabilities and SlowFast's spatiotemporal sequence modeling capabilities. Secondly, we introduce the confidence maximum heap to prune the anchors detected in each frame of the picture to filter out the effective anchors. These designs enable our SFMViT to achieve a mAP of 26.62% in the Chaotic World dataset, far exceeding existing models. Code is available at https://github.com/jfightyr/SlowFast-Meet-ViT.

You Wu, Yongxin Li, Mengyuan Liu et al.

Transformer-based models have improved visual tracking, but most still cannot run in real time on resource-limited devices, especially for unmanned aerial vehicle (UAV) tracking. To achieve a better balance between performance and efficiency, we propose AVTrack, an adaptive computation tracking framework that adaptively activates transformer blocks through an Activation Module (AM), which dynamically optimizes the ViT architecture by selectively engaging relevant components. To address extreme viewpoint variations, we propose to learn view-invariant representations via mutual information (MI) maximization. In addition, we propose AVTrack-MD, an enhanced tracker incorporating a novel MI maximization-based multi-teacher knowledge distillation framework. Leveraging multiple off-the-shelf AVTrack models as teachers, we maximize the MI between their aggregated softened features and the corresponding softened feature of the student model, improving the generalization and performance of the student, especially under noisy conditions. Extensive experiments show that AVTrack-MD achieves performance comparable to AVTrack's performance while reducing model complexity and boosting average tracking speed by over 17\%. Codes is available at: https://github.com/wuyou3474/AVTrack.

Ti Wang, Mengyuan Liu, Hong Liu et al.

Although data-driven methods have achieved success in 3D human pose estimation, they often suffer from domain gaps and exhibit limited generalization. In contrast, optimization-based methods excel in fine-tuning for specific cases but are generally inferior to data-driven methods in overall performance. We observe that previous optimization-based methods commonly rely on a projection constraint, which only ensures alignment in 2D space, potentially leading to the overfitting problem. To address this, we propose an Uncertainty-Aware testing-time Optimization (UAO) framework, which keeps the prior information of the pre-trained model and alleviates the overfitting problem using the uncertainty of joints. Specifically, during the training phase, we design an effective 2D-to-3D network for estimating the corresponding 3D pose while quantifying the uncertainty of each 3D joint. For optimization during testing, the proposed optimization framework freezes the pre-trained model and optimizes only a latent state. Projection loss is then employed to ensure the generated poses are well aligned in 2D space for high-quality optimization. Furthermore, we utilize the uncertainty of each joint to determine how much each joint is allowed for optimization. The effectiveness and superiority of the proposed framework are validated through extensive experiments on challenging datasets: Human3.6M, MPI-INF-3DHP, and 3DPW. Notably, our approach outperforms the previous best result by a large margin of 5.5\% on Human3.6M. Code is available at \href{https://github.com/xiu-cs/UAO-Pose3D}{https://github.com/xiu-cs/UAO-Pose3D}.

Mengyuan Liu, Jiajie Liu, Jinyan Zhang et al.

The lifting-based methods have dominated monocular 3D human pose estimation by leveraging detected 2D poses as intermediate representations. The 2D component of the final 3D human pose benefits from the detected 2D poses, whereas its depth counterpart must be estimated from scratch. The lifting-based methods encode the detected 2D pose and unknown depth in an entangled feature space, explicitly introducing depth uncertainty to the detected 2D pose, thereby limiting overall estimation accuracy. This work reveals that the depth representation is pivotal for the estimation process. Specifically, when depth is in an initial, completely unknown state, jointly encoding depth features with 2D pose features is detrimental to the estimation process. In contrast, when depth is initially refined to a more dependable state via network-based estimation, encoding it together with 2D pose information is beneficial. To address this limitation, we present a Mixture-of-Experts network for monocular 3D pose estimation named PoseMoE. Our approach introduces: (1) A mixture-of-experts network where specialized expert modules refine the well-detected 2D pose features and learn the depth features. This mixture-of-experts design disentangles the feature encoding process for 2D pose and depth, therefore reducing the explicit influence of uncertain depth features on 2D pose features. (2) A cross-expert knowledge aggregation module is proposed to aggregate cross-expert spatio-temporal contextual information. This step enhances features through bidirectional mapping between 2D pose and depth. Extensive experiments show that our proposed PoseMoE outperforms the conventional lifting-based methods on three widely used datasets: Human3.6M, MPI-INF-3DHP, and 3DPW.

Jiaying Lin, Shiman Wu, Jinfu Liu et al.

Millimeter-wave (mmWave) radar provides privacy-preserving sensing and is valuable for human action recognition (HAR). Existing mmWave point cloud datasets are limited in scale and mostly collected under homogeneous single-source settings, preventing current methods from handling real-world distribution shifts caused by heterogeneous radar sources, such as different devices and frequency bands. To address this, we introduce UniMM-HAR, the largest and first mmWave point cloud HAR dataset for heterogeneous multi-source scenarios, standardizing three distinct radar configurations to realistically evaluate cross-source generalization. We further propose the Doppler-aware Point Cloud Network (DAP-Net) to tackle heterogeneity challenges. DAP-Net enhances intra-modal representations and performs cross-modal alignment to learn source-invariant action semantics. Leveraging action-consistent spatio-temporal Doppler patterns as anchors, the Dual-space Doppler Reparameterization (D2R) module performs sample-adaptive geometric densification and Doppler-guided feature recalibration, while the Text Alignment Module (TAM) provides stable semantic anchors via a pretrained textual space. Experiments show that DAP-Net significantly outperforms existing methods under heterogeneous radar settings, achieving state-of-the-art accuracy and strong cross-source robustness.

Ziyi Wang, Peiming Li, Xinshun Wang et al.

Multimodal large language models (MLLMs) exhibit strong visual-language reasoning, yet remain confined to their native modalities and cannot directly process structured, non-visual data such as human skeletons. Existing methods either compress skeleton dynamics into lossy feature vectors for text alignment, or quantize motion into discrete tokens that generalize poorly across heterogeneous skeleton formats. We present SkeletonLLM, which achieves universal skeleton understanding by translating arbitrary skeleton sequences into the MLLM's native visual modality. At its core is DrAction, a differentiable, format-agnostic renderer that converts skeletal kinematics into compact image sequences. Because the pipeline is end-to-end differentiable, MLLM gradients can directly guide the rendering to produce task-informative visual tokens. To further enhance reasoning capabilities, we introduce a cooperative training strategy: Causal Reasoning Distillation transfers structured, step-by-step reasoning from a teacher model, while Discriminative Finetuning sharpens decision boundaries between confusable actions. SkeletonLLM demonstrates strong generalization on diverse tasks including recognition, captioning, reasoning, and cross-format transfer -- suggesting a viable path for applying MLLMs to non-native modalities. Code will be released upon acceptance.

Yiheng Liu, Hao He, Tianle Han et al.

The introduction of ChatGPT has led to a significant increase in the utilization of Large Language Models (LLMs) for addressing downstream tasks. There's an increasing focus on cost-efficient training and deployment within this context. Low-cost training and deployment of LLMs represent the future development trend. This paper reviews the evolution of large language model training techniques and inference deployment technologies aligned with this emerging trend. The discussion on training includes various aspects, including data preprocessing, training architecture, pre-training tasks, parallel training, and relevant content related to model fine-tuning. On the inference side, the paper covers topics such as model compression, parallel computation, memory scheduling, and structural optimization. It also explores LLMs' utilization and provides insights into their future development.

Wenhao Li, Mengyuan Liu, Hong Liu et al.

Transformers have been successfully applied in the field of video-based 3D human pose estimation. However, the high computational costs of these video pose transformers (VPTs) make them impractical on resource-constrained devices. In this paper, we present a hierarchical plug-and-play pruning-and-recovering framework, called Hierarchical Hourglass Tokenizer (H$_{2}$OT), for efficient transformer-based 3D human pose estimation from videos. H$_{2}$OT begins with progressively pruning pose tokens of redundant frames and ends with recovering full-length sequences, resulting in a few pose tokens in the intermediate transformer blocks and thus improving the model efficiency. It works with two key modules, namely, a Token Pruning Module (TPM) and a Token Recovering Module (TRM). TPM dynamically selects a few representative tokens to eliminate the redundancy of video frames, while TRM restores the detailed spatio-temporal information based on the selected tokens, thereby expanding the network output to the original full-length temporal resolution for fast inference. Our method is general-purpose: it can be easily incorporated into common VPT models on both seq2seq and seq2frame pipelines while effectively accommodating different token pruning and recovery strategies. In addition, our H$_{2}$OT reveals that maintaining the full pose sequence is unnecessary, and a few pose tokens of representative frames can achieve both high efficiency and estimation accuracy. Extensive experiments on multiple benchmark datasets demonstrate both the effectiveness and efficiency of the proposed method. Code and models are available at https://github.com/NationalGAILab/HoT.

Peiming Li, Ziyi Wang, Yulin Yuan et al.

Point cloud videos capture dynamic 3D motion while reducing the effects of lighting and viewpoint variations, making them highly effective for recognizing subtle and continuous human actions. Although Selective State Space Models (SSMs) have shown good performance in sequence modeling with linear complexity, the spatio-temporal disorder of point cloud videos hinders their unidirectional modeling when directly unfolding the point cloud video into a 1D sequence through temporally sequential scanning. To address this challenge, we propose the Unified Spatio-Temporal State Space Model (UST-SSM), which extends the latest advancements in SSMs to point cloud videos. Specifically, we introduce Spatial-Temporal Selection Scanning (STSS), which reorganizes unordered points into semantic-aware sequences through prompt-guided clustering, thereby enabling the effective utilization of points that are spatially and temporally distant yet similar within the sequence. For missing 4D geometric and motion details, Spatio-Temporal Structure Aggregation (STSA) aggregates spatio-temporal features and compensates. To improve temporal interaction within the sampled sequence, Temporal Interaction Sampling (TIS) enhances fine-grained temporal dependencies through non-anchor frame utilization and expanded receptive fields. Experimental results on the MSR-Action3D, NTU RGB+D, and Synthia 4D datasets validate the effectiveness of our method. Our code is available at https://github.com/wangzy01/UST-SSM.

Mengyuan Liu, Xinshun Wang, Zhongbin Fang et al.

This paper aims to model 3D human motion across domains, where a single model is expected to handle multiple modalities, tasks, and datasets. Existing cross-domain models often rely on domain-specific components and multi-stage training, which limits their practicality and scalability. To overcome these challenges, we propose a new setting to train a unified cross-domain model through a single process, eliminating the need for domain-specific components and multi-stage training. We first introduce Pose-in-Context (PiC), which leverages in-context learning to create a pose-centric cross-domain model. While PiC generalizes across multiple pose-based tasks and datasets, it encounters difficulties with modality diversity, prompting strategy, and contextual dependency handling. We thus propose Human-in-Context (HiC), an extension of PiC that broadens generalization across modalities, tasks, and datasets. HiC combines pose and mesh representations within a unified framework, expands task coverage, and incorporates larger-scale datasets. Additionally, HiC introduces a max-min similarity prompt sampling strategy to enhance generalization across diverse domains and a network architecture with dual-branch context injection for improved handling of contextual dependencies. Extensive experimental results show that HiC performs better than PiC in terms of generalization, data scale, and performance across a wide range of domains. These results demonstrate the potential of HiC for building a unified cross-domain 3D human motion model with improved flexibility and scalability. The source codes and models are available at https://github.com/BradleyWang0416/Human-in-Context.

Bin Ren, Xiaoshui Huang, Mengyuan Liu et al.

Vision transformers (ViTs) have recently been widely applied to 3D point cloud understanding, with masked autoencoding as the predominant pre-training paradigm. However, the challenge of learning dense and informative semantic features from point clouds via standard ViTs remains underexplored. We propose MaskClu, a novel unsupervised pre-training method for ViTs on 3D point clouds that integrates masked point modeling with clustering-based learning. MaskClu is designed to reconstruct both cluster assignments and cluster centers from masked point clouds, thus encouraging the model to capture dense semantic information. Additionally, we introduce a global contrastive learning mechanism that enhances instance-level feature learning by contrasting different masked views of the same point cloud. By jointly optimizing these complementary objectives, i.e., dense semantic reconstruction, and instance-level contrastive learning. MaskClu enables ViTs to learn richer and more semantically meaningful representations from 3D point clouds. We validate the effectiveness of our method via multiple 3D tasks, including part segmentation, semantic segmentation, object detection, and classification, where MaskClu sets new competitive results. The code and models will be released at:https://github.com/Amazingren/maskclu.

Ziyi Wang, Peiming Li, Hong Liu et al.

Natural Human-Robot Interaction (N-HRI) requires robots to recognize human actions at varying distances and states, regardless of whether the robot itself is in motion or stationary. This setup is more flexible and practical than conventional human action recognition tasks. However, existing benchmarks designed for traditional action recognition fail to address the unique complexities in N-HRI due to limited data, modalities, task categories, and diversity of subjects and environments. To address these challenges, we introduce ACTIVE (Action from Robotic View), a large-scale dataset tailored specifically for perception-centric robotic views prevalent in mobile service robots. ACTIVE comprises 30 composite action categories, 80 participants, and 46,868 annotated video instances, covering both RGB and point cloud modalities. Participants performed various human actions in diverse environments at distances ranging from 3m to 50m, while the camera platform was also mobile, simulating real-world scenarios of robot perception with varying camera heights due to uneven ground. This comprehensive and challenging benchmark aims to advance action and attribute recognition research in N-HRI. Furthermore, we propose ACTIVE-PC, a method that accurately perceives human actions at long distances using Multilevel Neighborhood Sampling, Layered Recognizers, Elastic Ellipse Query, and precise decoupling of kinematic interference from human actions. Experimental results demonstrate the effectiveness of ACTIVE-PC. Our code is available at: https://github.com/wangzy01/ACTIVE-Action-from-Robotic-View.

Jialun Cai, Mengyuan Liu, Hong Liu et al.

The widespread application of 3D human pose estimation (HPE) is limited by resource-constrained edge devices, requiring more efficient models. A key approach to enhancing efficiency involves designing networks based on the structural characteristics of input data. However, effectively utilizing the structural priors in human skeletal inputs remains challenging. To address this, we leverage both explicit and implicit spatio-temporal priors of the human body through innovative model design and a pre-training proxy task. First, we propose a Nano Human Topology Network (NanoHTNet), a tiny 3D HPE network with stacked Hierarchical Mixers to capture explicit features. Specifically, the spatial Hierarchical Mixer efficiently learns the human physical topology across multiple semantic levels, while the temporal Hierarchical Mixer with discrete cosine transform and low-pass filtering captures local instantaneous movements and global action coherence. Moreover, Efficient Temporal-Spatial Tokenization (ETST) is introduced to enhance spatio-temporal interaction and reduce computational complexity significantly. Second, PoseCLR is proposed as a general pre-training method based on contrastive learning for 3D HPE, aimed at extracting implicit representations of human topology. By aligning 2D poses from diverse viewpoints in the proxy task, PoseCLR aids 3D HPE encoders like NanoHTNet in more effectively capturing the high-dimensional features of the human body, leading to further performance improvements. Extensive experiments verify that NanoHTNet with PoseCLR outperforms other state-of-the-art methods in efficiency, making it ideal for deployment on edge devices like the Jetson Nano. Code and models are available at https://github.com/vefalun/NanoHTNet.

Sheng Yan, Yang Liu, Haoqiang Wang et al.

Cross-modal retrieval of image-text and video-text is a prominent research area in computer vision and natural language processing. However, there has been insufficient attention given to cross-modal retrieval between human motion and text, despite its wide-ranging applicability. To address this gap, we utilize a concise yet effective dual-unimodal transformer encoder for tackling this task. Recognizing that overlapping atomic actions in different human motion sequences can lead to semantic conflicts between samples, we explore a novel triplet loss function called DropTriple Loss. This loss function discards false negative samples from the negative sample set and focuses on mining remaining genuinely hard negative samples for triplet training, thereby reducing violations they cause. We evaluate our model and approach on the HumanML3D and KIT Motion-Language datasets. On the latest HumanML3D dataset, we achieve a recall of 62.9% for motion retrieval and 71.5% for text retrieval (both based on R@10). The source code for our approach is publicly available at https://github.com/eanson023/rehamot.