Reproduce “FRIDU: Functional Map Refinement with Guided Image Diffusion” 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
Avigail Cohen Rimon, Mirela Ben-Chen, Or Litany
FRIDU refines functional maps for matching 3D shapes by treating compact matrices as images and denoising them with a conditional diffusion model. At inference, point-to-point maps recovered from intermediate estimates guide denoising, with optional geometric objectives. On FAUST-to-FAUST tests, the pipeline reports mean geodesic error ×100 of 1.7, versus 1.9 for DiffZO and 2.5 initially; cross-dataset superiority is not established.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.