Reproduce “Exo2EgoHOI: Hand-Object-Interaction Aware Exocentric-to-Egocentric Video Generation” 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
Hongjia Zhai, Xiyu Zhang, Haoran Zhang, Zhichao Ye, Haomin Liu, Guofeng Zhang, Ian Reid, Xingxing Zuo
Exo2EgoHOI translates third-person manipulation videos into first-person views using target-view scene, articulated-hand, and hand–object relation cues, plus separate object and background visual anchors. In a hinged-object example, it better preserves the raised lid and circular interior. On ARCTIC-HOI, object mIoU reaches 0.3881 and PA-MPJPE 10.18 mm, but not every metric improves across datasets.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.