Reproduce “Vision-Based 6-DoF Grasp Pose Estimation for Robot Cloth Unfolding” 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
Domen Tabernik, Peter Nimac, Jan Jerićević, Danijel Skočaj, Andrej Gams
CeDiRNet-6DoF predicts a cloth grasp location and gripper orientation from an image, then filters candidate poses for robot reachability, inverse-kinematics validity, and collision-free execution. Its full configuration combines 6-DoF prediction with segmentation supervision, background randomization, and workspace cropping. On combined towel and T-shirt evaluations, it reached 60.9% coverage and 68.8% grasp success, within those categories and setups.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.