Reproduce “Query-Conditioned Articulation Estimation from a Single Image” 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
Abdelrhman Werby, Fabio Scaparro, Kai O. Arras
QueryArt predicts the joint type and motion geometry of a selected point from a calibrated RGB image and its location. It represents revolute hinge offset relative to query depth, allowing a depth measurement to restore metric scale. On held-out HOI!, it achieves 69.08% success versus 58.58% for 3DOI; robot trials succeed in 40/57 under reported criteria, with grasp failures excluded.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.