Reproduce “Gauge Equivariant Mesh CNNs: Anisotropic convolutions on geometric graphs” 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

Pim de Haan, Maurice Weiler, Taco Cohen, Max Welling

GEM-CNN introduces convolutions that use edge directions while accounting for arbitrary local reference frames. It transports neighbor features into each vertex’s frame and combines them with anisotropic filters. On FAUST, it predicts registered vertex identities with 99.73 ± 0.04% accuracy, versus 1.40 ± 0.5% for the Graph CNN baseline, over three runs; other methods’ scores are from papers, not controlled comparisons.

Summary by alphaXiv.

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