GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimation

TL;DR

GTPT is a novel Transformer-based approach that efficiently performs human pose estimation by gradually introducing keypoints and pruning tokens, reducing computation while maintaining high accuracy, especially for whole-body keypoints.

Contribution

The paper introduces GTPT, a group-based token pruning Transformer that improves efficiency and performance in human pose estimation, particularly for complex whole-body keypoints.

Findings

GTPT achieves higher accuracy with less computation on COCO datasets.

GTPT effectively handles numerous keypoints in whole-body pose estimation.

The method outperforms existing approaches in efficiency and accuracy.

Abstract

In recent years, 2D human pose estimation has made significant progress on public benchmarks. However, many of these approaches face challenges of less applicability in the industrial community due to the large number of parametric quantities and computational overhead. Efficient human pose estimation remains a hurdle, especially for whole-body pose estimation with numerous keypoints. While most current methods for efficient human pose estimation primarily rely on CNNs, we propose the Group-based Token Pruning Transformer (GTPT) that fully harnesses the advantages of the Transformer. GTPT alleviates the computational burden by gradually introducing keypoints in a coarse-to-fine manner. It minimizes the computation overhead while ensuring high performance. Besides, GTPT groups keypoint tokens and prunes visual tokens to improve model performance while reducing redundancy. We propose the Multi-Head Group Attention (MHGA) between different groups to achieve global interaction with little computational overhead. We conducted experiments on COCO and COCO-WholeBody. Compared to other methods, the experimental results show that GTPT can achieve higher performance with less computation, especially in whole-body with numerous keypoints.