Preprint
Aug 2026
SC-Diff: Semantically Calibrated Diffusion for Visible-to-Infrared Image Translation
This work proposes SC-Diff, a semantically calibrated latent diffusion framework that uses semantic priors for both conditional guidance and internal self-attention calibration, and introduces Semantic-Guided Self-Attention Calibration, which adaptively applies positive biases to query-key pairs of the same category.
Junyin Zhang, Siyu Huang, Jianxiong Ye et al.
· 0 citations