Reproduce “HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jongbin Lim, Taeyun Ha, Seongho Cha, Kanghyeon Cho, Mingi Choi, Subin Jeon, Jisoo Kim, Byungjun Kim and 1 more
HRDexDB links human grasp references to teleoperated robot executions, allowing robot-specific changes in finger configuration and contact placement rather than assuming identical grasps. Retrieval-Assisted Grasping succeeded on 15/20 unseen objects (75%), versus 8/20 (40%) without retrieval, in a real-world test using the Allegro V5 hand; this 20-trial result does not establish performance across hands or tasks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.