Search papers, labs, and topics across Lattice.
2
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.
Scaling up avatar pre-training to 1M in-the-wild videos unlocks emergent generalization capabilities like relightability and garment support, even without direct supervision.
