Reproduce “AutoDex: An Automated Real-World System for Dexterous Grasping Data Collection” 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
Mingi Choi, Gunhee Kim, Jisoo Kim, Taeksoo Kim, Taeyun Ha, Jongbin Lim, Hanbyul Joo
AutoDex automates physical testing of generated dexterous grasp candidates by screening and executing them, labeling outcomes from camera tracking, and resetting objects. Across 100 objects, it collected 3,593 labeled trials with Allegro and Inspire hands. In a retrieval test on 20 objects and 515 trials, physically validated grasps succeeded 76% versus 34% for simulation-screened candidates, within tested conditions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.