Reproduce “CARI4D: Category Agnostic 4D Reconstruction of Human-Object Interaction” 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
Xianghui Xie, Bowen Wen, Yan Chang, Hesam Rabeti, Jiefeng Li, Ye Yuan, Gerard Pons-Moll, Stan Birchfield
CARI4D reconstructs human-object 3D motion, scale, and interaction from monocular video by aligning predictions from pretrained models and jointly refining the person and object with image evidence, motion constraints, and predicted contacts. On 62 selected BEHAVE videos, its combined-mesh Chamfer distance was 9.23 cm versus VisTracker’s 14.22 cm, with VisTracker given aligned CARI4D object meshes; first-frame visibility is assumed.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.