TriGait: Aligning and Fusing Skeleton and Silhouette Gait Data via a Tri-Branch Network

TL;DR

TriGait introduces a novel multi-branch network that effectively combines skeleton and silhouette data for gait recognition, significantly improving accuracy across various conditions.

Contribution

It proposes a triple-branch framework that aligns and fuses skeleton and silhouette features, enhancing gait recognition performance.

Findings

Achieves 96.0% rank-1 accuracy on CASIA-B dataset.

Outperforms state-of-the-art methods significantly.

Effective integration of multi-modal gait features.

Abstract

Gait recognition is a promising biometric technology for identification due to its non-invasiveness and long-distance. However, external variations such as clothing changes and viewpoint differences pose significant challenges to gait recognition. Silhouette-based methods preserve body shape but neglect internal structure information, while skeleton-based methods preserve structure information but omit appearance. To fully exploit the complementary nature of the two modalities, a novel triple branch gait recognition framework, TriGait, is proposed in this paper. It effectively integrates features from the skeleton and silhouette data in a hybrid fusion manner, including a two-stream network to extract static and motion features from appearance, a simple yet effective module named JSA-TC to capture dependencies between all joints, and a third branch for cross-modal learning by aligning and fusing low-level features of two modalities. Experimental results demonstrate the superiority and effectiveness of TriGait for gait recognition. The proposed method achieves a mean rank-1 accuracy of 96.0% over all conditions on CASIA-B dataset and 94.3% accuracy for CL, significantly outperforming all the state-of-the-art methods. The source code will be available at https://github.com/feng-xueling/TriGait/.