With the availability of egocentric 3D hand-object interaction datasets, there is increasing interest in developing unified models for hand-object pose estimation and action recognition. However, existing methods still struggle to recognise seen actions on unseen objects due to the limitations in representing object shape and movement using 3D bounding boxes. Additionally, the reliance on object templates at test time limits their generalisability to unseen objects. To address these challenges, we propose to leverage superquadrics as an alternative 3D object representation to bounding boxes and demonstrate their effectiveness on both template-free object reconstruction and action recognition tasks. Moreover, as we find that pure appearance-based methods can outperform the unified methods, the potential benefits from 3D geometric information remain unclear. Therefore, we study the compositionality of actions by considering a more challenging task where the training combinations of verbs and nouns do not overlap with the testing split. We extend H2O and FPHA datasets with compositional splits and design a novel collaborative learning framework that can explicitly reason about the geometric relations between hands and the manipulated object. Through extensive quantitative and qualitative evaluations, we demonstrate significant improvements over the state-of-the-arts in (compositional) action recognition.

Modeling high-resolution spatiotemporal representations, including both global dynamic contexts (e.g., holistic human motion tendencies) and local motion details (e.g., high-frequency changes of keypoints), is essential for video-based human pose estimation (VHPE). Current state-of-the-art methods typically unify spatiotemporal learning within a single type of modeling structure (convolution or attention-based blocks), which inherently have difficulties in balancing global and local dynamic modeling and may bias the network to one of them, leading to suboptimal performance. Moreover, existing VHPE models suffer from quadratic complexity when capturing global dependencies, limiting their applicability especially for high-resolution sequences. Recently, the state space models (known as Mamba) have demonstrated significant potential in modeling long-range contexts with linear complexity; however, they are restricted to 1D sequential data. In this paper, we present a novel framework that extends Mamba from two aspects to separately learn global and local high-resolution spatiotemporal representations for VHPE. Specifically, we first propose a Global Spatiotemporal Mamba, which performs 6D selective space-time scan and spatial- and temporal-modulated scan merging to efficiently extract global representations from high-resolution sequences. We further introduce a windowed space-time scan-based Local Refinement Mamba to enhance the high-frequency details of localized keypoint motions. Extensive experiments on four benchmark datasets demonstrate that the proposed model outperforms state-of-the-art VHPE approaches while achieving better computational trade-offs.

Human pose estimation in videos has long been a compelling yet challenging task within the realm of computer vision. Nevertheless, this task remains difficult because of the complex video scenes, such as video defocus and self-occlusion. Recent methods strive to integrate multi-frame visual features generated by a backbone network for pose estimation. However, they often ignore the useful joint information encoded in the initial heatmap, which is a by-product of the backbone generation. Comparatively, methods that attempt to refine the initial heatmap fail to consider any spatio-temporal motion features. As a result, the performance of existing methods for pose estimation falls short due to the lack of ability to leverage both local joint (heatmap) information and global motion (feature) dynamics.

Temporal modeling is crucial for multi-frame human pose estimation. Most existing methods directly employ optical flow or deformable convolution to predict full-spectrum motion fields, which might incur numerous irrelevant cues, such as a nearby person or background. Without further efforts to excavate meaningful motion priors, their results are suboptimal, especially in complicated spatio-temporal interactions. On the other hand, the temporal difference has the ability to encode representative motion information which can potentially be valuable for pose estimation but has not been fully exploited. In this paper, we present a novel multi-frame human pose estimation framework, which employs temporal differences across frames to model dynamic contexts and engages mutual information objectively to facilitate useful motion information disentanglement. To be specific, we design a multi-stage Temporal Difference Encoder that performs incremental cascaded learning conditioned on multi-stage feature difference sequences to derive informative motion representation. We further propose a Representation Disentanglement module from the mutual information perspective, which can grasp discriminative task-relevant motion signals by explicitly defining useful and noisy constituents of the raw motion features and minimizing their mutual information. These place us to rank No.1 in the Crowd Pose Estimation in Complex Events Challenge on benchmark dataset HiEve, and achieve state-of-the-art performance on three benchmarks PoseTrack2017, PoseTrack2018, and PoseTrack21.

Denoising diffusion probabilistic models that were initially proposed for realistic image generation have recently shown success in various perception tasks (e.g., object detection and image segmentation) and are increasingly gaining attention in computer vision. However, extending such models to multi-frame human pose estimation is non-trivial due to the presence of the additional temporal dimension in videos. More importantly, learning representations that focus on keypoint regions is crucial for accurate localization of human joints. Nevertheless, the adaptation of the diffusion-based methods remains unclear on how to achieve such objective. In this paper, we present DiffPose, a novel diffusion architecture that formulates video-based human pose estimation as a conditional heatmap generation problem. First, to better leverage temporal information, we propose SpatioTemporal Representation Learner which aggregates visual evidences across frames and uses the resulting features in each denoising step as a condition. In addition, we present a mechanism called Lookup-based Multi-Scale Feature Interaction that determines the correlations between local joints and global contexts across multiple scales. This mechanism generates delicate representations that focus on keypoint regions. Altogether, by extending diffusion models, we show two unique characteristics from DiffPose on pose estimation task: (i) the ability to combine multiple sets of pose estimates to improve prediction accuracy, particularly for challenging joints, and (ii) the ability to adjust the number of iterative steps for feature refinement without retraining the model. DiffPose sets new state-of-the-art results on three benchmarks: PoseTrack2017, PoseTrack2018, and PoseTrack21.

Multi-frame human pose estimation has long been an appealing and fundamental issue in visual perception. Owing to the frequent rapid motion and pose occlusion in videos, this task is extremely challenging. Current state-of-the-art methods seek to model spatiotemporal features by equally fusing each frame in the local sequence, which weakens the target frame information. In addition, existing approaches usually emphasize more on deep features while ignoring the detailed information implied in the shallow feature maps, resulting in the dropping of crucial features. To address the above problems, we propose an effective framework, namely spatiotemporal learning transformer for video-based human pose estimation (SLT-Pose), which consists of a Personalized Feature Extraction Module (PFEM), Self-feature Refinement Module (SRM), Cross-frame Temporal Learning Module (CTLM) and Disentangled Keypoint Detector (DKD). To be specific, we propose PFEM which extracts and modulates the individual frame features to adapt to the varying human shape, and integrates single-frame features to obtain the spatiotemporal features. We further present SRM to establish global correlation spatial cues on the target frame to attain the refinement feature. Then, a CTLM is designed to search for the information most closely related to the target frame from the spatiotemporal features to intensify the interaction between the target frame and the local sequence, using both the shallow detailed and the deep semantic representations. Finally, we employ DKD to extract the disentangled characteristics of each joint and encode the articulated joint pairs in the human body, promoting the model to reasonably and accurately predict the keypoint heatmaps. Extensive experiments on three huamn motion benchmarks, including PoseTrack2017, PoseTrack2018, and Sub-JHMDB dataset, demonstrate that SLT-Pose plays favorably against state-of-the-art approaches in terms of both objective evaluation and subjective visual performance.

Multi-frame human pose estimation has long been a compelling and fundamental problem in computer vision. This task is challenging due to fast motion and pose occlusion that frequently occur in videos. State-of-the-art methods strive to incorporate additional visual evidences from neighboring frames (supporting frames) to facilitate the pose estimation of the current frame (key frame). One aspect that has been obviated so far, is the fact that current methods directly aggregate unaligned contexts across frames. The spatial-misalignment between pose features of the current frame and neighboring frames might lead to unsatisfactory results. More importantly, existing approaches build upon the straightforward pose estimation loss, which unfortunately cannot constrain the network to fully leverage useful information from neighboring frames. To tackle these problems, we present a novel hierarchical alignment framework, which leverages coarse-to-fine deformations to progressively update a neighboring frame to align with the current frame at the feature level. We further propose to explicitly supervise the knowledge extraction from neighboring frames, guaranteeing that useful complementary cues are extracted. To achieve this goal, we theoretically analyzed the mutual information between the frames and arrived at a loss that maximizes the task-relevant mutual information. These allow us to rank No.1 in the Multi-frame Person Pose Estimation Challenge on benchmark dataset PoseTrack2017, and obtain state-of-the-art performance on benchmarks Sub-JHMDB and Pose-Track2018. Our code is released at https://github.com/Pose-Group/FAMI-Pose, hoping that it will be useful to the community.

Estimating human poses from a video is at the foundation of many visual intelligent systems. Various convolutional neural networks have been proposed, achieving state-of-the-art performance on different image datasets. However, most existing approaches are image based, which deliver unreliable estimations on videos since they fail to model temporal consistency across video frames. Recently, another line of work leverages temporal cues for multi-frame person pose estimation, yet still in an instance-unaware fashion, disregarding the specific traits of different instances (persons) or different joints. In this paper, we propose a novel approach to learn specific keypoint motion representations for each person, termed Personalized Motion-Aware Network (PMAN). In the PMAN, we devise three components: (i) an Instance-Sensitive Extractor that adaptively computes the spatial features according to human physical characteristics; (ii) a Keypoint Motion Encoder that separately generates convolution kernels with fine-grained keypoint motion encoding; (iii) a Motion Driven Decoder that parses multi-frame spatial features of the same person to provide precise human pose estimations. Extensive experiments on PoseTrack2017 and PoseTrack2018 datasets demonstrate that our approach greatly improves the performance of multi-frame human pose estimation. It is worth mentioning that our approach surpasses the state-of-the-art method by +1.7 mAP and achieves 82.9 mAP on PoseTrack2017 dataset.

Multi-frame human pose estimation is at the core of many computer vision tasks. Although state-of-the-art approaches have demonstrated remarkable results for human pose estimation on static images, their performances inevitably come short when being applied to videos. A central issue lies in the visual degeneration of video frames induced by rapid motion and pose occlusion in dynamic environments. This problem, by nature, is insurmountable for a single frame. Therefore, incorporating complementary visual cues from other video frames becomes an intuitive paradigm. Current state-of-the-art methods usually leverage information from adjacent frames, which unfortunately place excessive focus on only the temporally nearby frames. In this paper, we argue that combining global semantically similar information and local temporal visual context will deliver more comprehensive and more robust representations for human pose estimation. Towards this end, we present an effective framework, namely global-local enhanced pose estimation ( GLPose ) network. Our framework consists of a feature processing module that conditionally incorporates global semantic information and local visual context to generate a robust human representation and a feature enhancement module that excavates complementary information from this aggregated representation to enhance keyframe features for precise estimation. We empirically find that the proposed GLpose outperforms existing methods by a large margin and achieves new state-of-the-art results on large benchmark datasets.

Multi-frame human pose estimation in complicated situations is challenging. Although state-of-the-art human joints detectors have demonstrated remarkable results for static images, their performances come short when we apply these models to video sequences. Prevalent shortcomings include the failure to handle motion blur, video defocus, or pose occlusions, arising from the inability in capturing the temporal dependency among video frames. On the other hand, directly employing conventional recurrent neural networks incurs empirical difficulties in modeling spatial contexts, especially for dealing with pose occlusions. In this paper, we propose a novel multi-frame human pose estimation framework, leveraging abundant temporal cues between video frames to facilitate keypoint detection. Three modular components are designed in our framework. A Pose Temporal Merger encodes keypoint spatiotemporal context to generate effective searching scopes while a Pose Residual Fusion module computes weighted pose residuals in dual directions. These are then processed via our Pose Correction Network for efficient refining of pose estimations. Our method ranks No.1 in the Multi-frame Person Pose Estimation Challenge on the large-scale benchmark datasets PoseTrack2017 and PoseTrack2018. We have released our code, hoping to inspire future research.