Reproduce “Pose-based Neural Clothing for Animated Characters” 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

Julian N. Heidenreich, Vinicius C. Azevedo, Jakob Buhmann, Lento Manickathan, Arnold Moon, Paul Kanyuk, Amit H. Bermano, Jingwei Tang

The authors present a learned per-frame garment deformer that predicts clothing on a posed character without running a full simulation for each preview. It learns from production simulations and artist-approved examples using a mesh-aware variational autoencoder, pose and body-surface cues, and a wrinkle-orientation loss. On held-out JEN shirt poses, MAE is 0.3605 versus 0.4202 for an adapted per-joint MLP; scope is single-garment, quasi-static cloth, not history-dependent dynamics.

Summary by alphaXiv.

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