3D Human Pose Estimation via Spatial Graph Order Attention and Temporal Body Aware Transformer

TL;DR

This paper introduces a novel 3D human pose estimation method combining graph order attention and a body-aware transformer to better model spatial and temporal dependencies, achieving state-of-the-art results.

Contribution

It proposes a new graph modeling approach with multiple graph orders and a graph order attention module, along with a temporal body-aware transformer that incorporates pose centrality.

Findings

Effective on Human3.6m, MPIINF-3DHP, and HumanEva-I datasets.

Outperforms existing methods in 3D pose estimation accuracy.

Code and models are publicly available.

Abstract

Nowadays, Transformers and Graph Convolutional Networks (GCNs) are the prevailing techniques for 3D human pose estimation. However, Transformer-based methods either ignore the spatial neighborhood relationships between the joints when used for skeleton representations or disregard the local temporal patterns of the local joint movements in skeleton sequence modeling, while GCN-based methods often neglect the need for pose-specific representations. To address these problems, we propose a new method that exploits the graph modeling capability of GCN to represent each skeleton with multiple graphs of different orders, incorporated with a newly introduced Graph Order Attention module that dynamically emphasizes the most representative orders for each joint. The resulting spatial features of the sequence are further processed using a proposed temporal Body Aware Transformer that models the global body feature dependencies in the sequence with awareness of the local inter-skeleton feature dependencies of joints. Given that our 3D pose output aligns with the central 2D pose in the sequence, we improve the self-attention mechanism to be aware of the central pose while diminishing its focus gradually towards the first and the last poses. Extensive experiments on Human3.6m, MPIINF-3DHP, and HumanEva-I datasets demonstrate the effectiveness of the proposed method. Code and models are made available on Github.