Hand-Object Interaction Reconstruction from Video via Agentic Generation

Jin-Chuan Shi^1,*, Binhong Ye^1,*, Tao Liu¹, Junzhe He¹, Yangjinhui Xu¹, Xiaoyang Liu¹, Zeju Li¹, Hao Chen¹, Chunhua Shen^1,2

¹ State Key Lab of CAD & CG, Zhejiang University · ² Zhejiang University of Technology · * Equal contribution

Scan for paper, code
& project page

agile-hoi.github.io · best viewed fullscreen

Reconstructed · Monocular Video

—

◎ Drag to explore · tap a sequence

Initializing… 0%

Frame 001 / 0

Video overlay 35%