HUGS: Guiding Unified Dexterous Grasp Synthesis Across Modes and Scales via Learned Human Priors
Learns an object-conditioned human grasp prior from 1.8K human grasps to pick contact modes and wrist starts for force-closure optimisation; synthesises 3.2M grasps from pinches to bimanual, 2–30 cm objects.从 1800 次人类抓取中学习物体条件的人类先验，用来选择接触模式和手腕初值以指导力封闭优化；合成 320 万个抓取，覆盖两指捏取到双手抓取、2–30 cm 物体。
BibTeX
@misc{yu2026hugs,
title = {{HUGS: Guiding Unified Dexterous Grasp Synthesis Across Modes and Scales via Learned Human Priors}},
author = {Mingrui Yu and Yongpeng Jiang and Yongyi Jia and Kangchen Lv and Li Huang and Yi Ren and Xiang Li},
year = {2026},
eprint = {2607.04554},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2607.04554},
}Manipulation & Dexterous Hands操作与灵巧手Robot arm机械臂Dexterous hand灵巧手dexterous-graspinghuman-priorgrasp-synthesisbimanualreal