Reproduce “Putting People in their Place: Monocular Regression of 3D People in Depth” 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

Yu Sun, Wu Liu, Qian Bao, Yili Fu, Tao Mei, Michael J. Black

BEV estimates 3D poses, shapes, and depth relations for multiple people from one image. It combines the ordinary view with an inferred overhead map, using relative-depth and age-group labels to guide reconstruction. On RH, it achieves 68.27% PCDR versus CRMH’s 54.83%; on AGORA kids, its all-set NMVE is 228.9 versus ROMP’s 313.2. These benchmark results do not resolve monocular ambiguity.

Summary by alphaXiv.

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