UGTSR improves codebook-based blind super-resolution by combining uncertainty-guided loss weighting, top-3 codebook matching, and an align-attention module for fusing low- and high-quality features.
Swinir: Image restoration using swin transformer,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Incorporating Uncertainty-Guided and Top-k Codebook Matching for Real-World Blind Image Super-Resolution
UGTSR improves codebook-based blind super-resolution by combining uncertainty-guided loss weighting, top-3 codebook matching, and an align-attention module for fusing low- and high-quality features.