A diffusion U-Net's attention features, aligned with a bidirectional attention score, can rank visual similarity competitively with CLIP, DINO, and LPIPS.
Emerg- ing properties in self-supervised vision transformers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
DiffSim: Taming Diffusion Models for Evaluating Visual Similarity
A diffusion U-Net's attention features, aligned with a bidirectional attention score, can rank visual similarity competitively with CLIP, DINO, and LPIPS.