Reproduce “PhyVGGT: Feed-Forward Articulated Object Reconstruction from Monocular Videos” 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

Qi Shao, Zhihao Cao, Shuhao Zhai, Anh Nguyen, Hong Liang, Baoru Huang, Liang Li

PhyVGGT reconstructs articulated objects from uncalibrated monocular RGB video. It identifies movable parts, aligns their motion to a stationary base, and restricts joint axes to the base’s three orthogonal directions. On Arti4D, joint accuracy within 10° is 60.4%, versus 36.9% for FAMOS. The method is designed for manufactured objects whose joints align with right-angle base geometry.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.