Kinematic-aware Hierarchical Attention Network for Human Pose Estimation in Videos

TL;DR

This paper introduces a kinematic-aware hierarchical transformer network that improves human pose estimation in videos by capturing keypoint velocities and accelerations, refining pose estimates, and handling occlusions more effectively.

Contribution

It proposes a novel architecture that leverages keypoint kinematic features and hierarchical transformers for enhanced spatio-temporal pose estimation.

Findings

Improved accuracy in 2D and 3D pose estimation tasks.

Effective handling of occlusions and motion jitter.

Versatile application across multiple human pose estimation tasks.

Abstract

Previous video-based human pose estimation methods have shown promising results by leveraging aggregated features of consecutive frames. However, most approaches compromise accuracy to mitigate jitter or do not sufficiently comprehend the temporal aspects of human motion. Furthermore, occlusion increases uncertainty between consecutive frames, which results in unsmooth results. To address these issues, we design an architecture that exploits the keypoint kinematic features with the following components. First, we effectively capture the temporal features by leveraging individual keypoint's velocity and acceleration. Second, the proposed hierarchical transformer encoder aggregates spatio-temporal dependencies and refines the 2D or 3D input pose estimated from existing estimators. Finally, we provide an online cross-supervision between the refined input pose generated from the encoder and the final pose from our decoder to enable joint optimization. We demonstrate comprehensive results and validate the effectiveness of our model in various tasks: 2D pose estimation, 3D pose estimation, body mesh recovery, and sparsely annotated multi-human pose estimation. Our code is available at https://github.com/KyungMinJin/HANet.