Reproduce “Identity World Model: Recognizing People in an Imagined Latent World” 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

Dingqiang Ye, Kartik Narayan, Vishal M. Patel

IDWorld predicts feature representations for five canonical conditions from RGB images, then weights them for each probe–gallery pair to compare people across clothing, viewpoint, and pose changes. On CCPG, latent imagination reaches 90.1% mean Rank-1 versus 90.6% for explicit imagination, while reducing inference latency from 67.1 to 5.9 ms per frame. Evaluation centers on gait and video-based ReID benchmarks.

Summary by alphaXiv.

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