English

Multiple View Geometry Transformers for 3D Human Pose Estimation

Computer Vision and Pattern Recognition 2023-11-21 v1 Artificial Intelligence Machine Learning

Abstract

In this work, we aim to improve the 3D reasoning ability of Transformers in multi-view 3D human pose estimation. Recent works have focused on end-to-end learning-based transformer designs, which struggle to resolve geometric information accurately, particularly during occlusion. Instead, we propose a novel hybrid model, MVGFormer, which has a series of geometric and appearance modules organized in an iterative manner. The geometry modules are learning-free and handle all viewpoint-dependent 3D tasks geometrically which notably improves the model's generalization ability. The appearance modules are learnable and are dedicated to estimating 2D poses from image signals end-to-end which enables them to achieve accurate estimates even when occlusion occurs, leading to a model that is both accurate and generalizable to new cameras and geometries. We evaluate our approach for both in-domain and out-of-domain settings, where our model consistently outperforms state-of-the-art methods, and especially does so by a significant margin in the out-of-domain setting. We will release the code and models: https://github.com/XunshanMan/MVGFormer.

Keywords

Cite

@article{arxiv.2311.10983,
  title  = {Multiple View Geometry Transformers for 3D Human Pose Estimation},
  author = {Ziwei Liao and Jialiang Zhu and Chunyu Wang and Han Hu and Steven L. Waslander},
  journal= {arXiv preprint arXiv:2311.10983},
  year   = {2023}
}

Comments

14 pages, 8 figures

R2 v1 2026-06-28T13:24:54.697Z