Search papers, labs, and topics across Lattice.
University of Electronic Science and Technology
1
0
3
Future-L1 shows that preserving visual semantics in latent space can dramatically enhance video event prediction accuracy, outperforming previous models by substantial margins.
