PoseGRAF: Geometric-Reinforced Adaptive Fusion for Monocular 3D Human Pose Estimation

TL;DR

PoseGRAF introduces a novel framework for monocular 3D human pose estimation that effectively models joint and bone dependencies using graph convolution, cross-attention, and adaptive fusion, outperforming existing methods on standard datasets.

Contribution

The paper presents a dual graph convolutional structure with cross-attention and dynamic fusion modules, along with an improved Transformer encoder, to enhance 3D pose estimation accuracy and robustness.

Findings

Outperforms state-of-the-art on Human3.6M and MPI-INF-3DHP datasets.

Demonstrates strong generalization on in-the-wild videos.

Effectively captures joint and bone dependencies for plausible pose estimation.

Abstract

Existing monocular 3D pose estimation methods primarily rely on joint positional features, while overlooking intrinsic directional and angular correlations within the skeleton. As a result, they often produce implausible poses under joint occlusions or rapid motion changes. To address these challenges, we propose the PoseGRAF framework. We first construct a dual graph convolutional structure that separately processes joint and bone graphs, effectively capturing their local dependencies. A Cross-Attention module is then introduced to model interdependencies between bone directions and joint features. Building upon this, a dynamic fusion module is designed to adaptively integrate both feature types by leveraging the relational dependencies between joints and bones. An improved Transformer encoder is further incorporated in a residual manner to generate the final output. Experimental results on the Human3.6M and MPI-INF-3DHP datasets show that our method exceeds state-of-the-art approaches. Additional evaluations on in-the-wild videos further validate its generalizability. The code is publicly available at https://github.com/iCityLab/PoseGRAF.