English

HandX: Scaling Bimanual Motion and Interaction Generation

Computer Vision and Pattern Recognition 2026-03-31 v1

Abstract

Synthesizing human motion has advanced rapidly, yet realistic hand motion and bimanual interaction remain underexplored. Whole-body models often miss the fine-grained cues that drive dexterous behavior, finger articulation, contact timing, and inter-hand coordination, and existing resources lack high-fidelity bimanual sequences that capture nuanced finger dynamics and collaboration. To fill this gap, we present HandX, a unified foundation spanning data, annotation, and evaluation. We consolidate and filter existing datasets for quality, and collect a new motion-capture dataset targeting underrepresented bimanual interactions with detailed finger dynamics. For scalable annotation, we introduce a decoupled strategy that extracts representative motion features, e.g., contact events and finger flexion, and then leverages reasoning from large language models to produce fine-grained, semantically rich descriptions aligned with these features. Building on the resulting data and annotations, we benchmark diffusion and autoregressive models with versatile conditioning modes. Experiments demonstrate high-quality dexterous motion generation, supported by our newly proposed hand-focused metrics. We further observe clear scaling trends: larger models trained on larger, higher-quality datasets produce more semantically coherent bimanual motion. Our dataset is released to support future research.

Keywords

Cite

@article{arxiv.2603.28766,
  title  = {HandX: Scaling Bimanual Motion and Interaction Generation},
  author = {Zimu Zhang and Yucheng Zhang and Xiyan Xu and Ziyin Wang and Sirui Xu and Kai Zhou and Bing Zhou and Chuan Guo and Jian Wang and Yu-Xiong Wang and Liang-Yan Gui},
  journal= {arXiv preprint arXiv:2603.28766},
  year   = {2026}
}

Comments

CVPR 2026. Project Page: https://handx-project.github.io. Code: https://github.com/handx-project/HandX

R2 v1 2026-07-01T11:44:36.776Z