Search papers, labs, and topics across Lattice.
D map that represents a distortion-free, D geometry and aspect ratio of the subject. From this normalized
4
0
4
Jointly modeling 3D geometry and relighting in a diffusion framework unlocks physically plausible single-image relighting that surpasses previous pipeline-based or geometry-agnostic approaches.
Achieve realistic human-scene interaction 50x faster by amortizing geometry-based fitting into a feed-forward transformer that iteratively refines human meshes.
Imagine training embodied agents with a dataset so realistic, humans prefer it 78% of the time – InHabit makes that possible.
Reconstructing complete, animatable 3D avatars from heavily occluded YouTube videos is now possible, thanks to a hallucination-as-supervision pipeline using diffusion models.
