Search papers, labs, and topics across Lattice.
1
0
3
17
Turns out, your pre-trained face recognition ViT already knows which faces are high quality, just by looking at the attention maps.
