UDAIR combines codebook quantization, cross-sample contrastive learning, and CORAL-based test-time adaptation to reduce the domain gap in all-in-one image restoration.
Bridg- ing the gap between haze scenarios: A unified image dehaz- ing model,
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
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
From Controlled Scenarios to Real-World: Cross-Domain Degradation Pattern Matching for All-in-One Image Restoration
UDAIR combines codebook quantization, cross-sample contrastive learning, and CORAL-based test-time adaptation to reduce the domain gap in all-in-one image restoration.