Reproduce “FlowRig: Unified Skeleton Generation for 3D and 4D Meshes via Flow Matching” 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
Fengzhi Guo, Nate Morrill, Shiqi Li, Cheng Zhang
FlowRig generates coordinated skeleton sequences for changing 4D meshes with a feedforward model that jointly processes frames and bone slots, and also supports static 3D rigging. On TexVerse full-observation reconstruction at 24 frames, it reports lower joint and bone errors than fine-tuned Puppeteer-D*, with zero topology-change scores; partial-observation prediction remains qualitatively evaluated.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.