Monocular video human mesh recovery is essential for digital humans, avatar animation, and embodied simulation, where both temporal stability and expressive whole-body motion are required. Existing video HMR methods produce coherent body motion but often overlook detailed hand articulation, while image-based whole-body methods recover SMPL-X meshes independently per frame, often leading to jittery and inaccurate hand motion. We present a temporally coherent whole-body HMR framework for challenging in-the-wild monocular videos. Our model unifies body context and part-specific hand observations through residual body-hand fusion, enabling stable body motion and detailed hand recovery within a single temporal architecture. We further introduce close-up-aware augmentation to improve robustness under upper-body framing. Experiments on whole-body and body-only benchmarks demonstrate improved hand reconstruction and competitive body accuracy. Our method also produces temporally stable and 2D-consistent SMPL-X motion in challenging real-world videos.
@article{arxiv.2605.18102,
title = {DanceHMR: Hand-Aware Whole-Body Human Mesh Recovery from Monocular Videos},
author = {Wenhao Shen and Ming Zhou and Hengyuan Zhang and Siyuan Bian and Youjiang Xu and Xi Lin},
journal= {arXiv preprint arXiv:2605.18102},
year = {2026}
}
Comments
I would like to withdraw my arXiv paper submission due to company-related approval and authorization requirements