Search papers, labs, and topics across Lattice.
1
0
2
4
Video diffusion models already contain implicit multi-view knowledge, making them surprisingly effective for novel view synthesis when adapted to ignore temporal coherence.
