Search papers, labs, and topics across Lattice.
1
0
2
11
Injecting semantic knowledge from vision transformer embeddings into guided depth super-resolution models dramatically reduces artifacts and sharpens depth boundaries.
