Reproduce “Deep Differentiable Grasp Planner for High-DOF Grippers” 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
Min Liu, Zherong Pan, Kai Xu, Kanishka Ganguly, Dinesh Manocha
The authors train a network to predict dexterous-hand grasp poses from depth images, using differentiable grasp-quality and collision feedback for inexact contacts. A generalized Q1 objective guides pose learning directly, avoiding runtime search over hand configurations. On 50 unseen YCB objects, mean Q1 was 0.23 and lift-success rate 66%, versus 0.11 and 54% for a prior predictor; matched test conditions and the combined approach limit this comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.