Reproduce “StyleFields: Multi-Scale AdaIN-Modulated Implicit SDFs for Coarse-to-Fine 3D Shape Reconstruction and Editing” 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

Ehsan Garaaghaji, Nicolas Talabot, Pascal Fua, Doruk Oner

StyleFields enables 3D shape reconstruction and scale-selective editing by replacing a single entangled shape code with four style codes assigned to different decoder depths. The model trains these depths with progressively sharper signed-distance targets, aligning shallow codes with coarse geometry and deeper codes with finer detail. Car examples show coarse-to-fine specialization, not perfectly independent controls; reconstruction performance varies across benchmarks.

Summary by alphaXiv.

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