Search papers, labs, and topics across Lattice.
Harbin Institute of Technology
1
0
3
4
Foundation models can be tamed to reconstruct realistic 4D interactions between hands and articulated objects from a single RGB video, even without pre-scanning or multi-view data.
