DDTNet: Degradation Disentanglement and Transfer Network for Test-Time All-in-One De-weathering Adaptation
2026-06-15 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors focus on fixing pictures taken in bad weather like rain, snow, or fog using one single tool. They noticed that it’s easier to model the weather problems in images than to try to instantly fix them. So, they made a system called DDTNet that learns the patterns of weather damage from certain images and applies these patterns to clean images to create custom practice data. This helps improve the tool’s ability to fix images from different weather types and places. Their tests show that this method helps existing all-in-one weather fixing tools work better on real-world images.
adverse weather image restorationdegradation patternsdomain adaptationdegradation disentanglementattention mechanismsimage derainingimage desnowingimage dehazingpaired training datafine-tuning
Authors
Kuan-Hung Lin, Fu-Jen Tsai, Yan-Tsung Peng, Min-Hung Chen, Chia-Wen Lin, Yen-Yu Lin
Abstract
All-in-one adverse weather image restoration aims to remove multiple degradations, such as rain, haze, and snow, using a single unified model. Despite their broad applicability, existing methods typically compromise performance, delivering balanced but suboptimal results for individual degradation types. This issue becomes more pronounced when a domain gap exists between training and testing data. Motivated by the observation that modeling degradation patterns is more feasible than recovering clean content, we propose the Degradation Disentanglement and Transfer Network (DDTNet), which focuses specifically on degradation transfer. By disentangling degradation patterns from target-domain degraded images and transferring them to source domain clean images, DDTNet generates domain-adaptive paired training data. These pairs are then used to fine-tune restoration models, significantly enhancing their adaptability across diverse weather conditions and domains. The core of DDTNet is the Degradation Disentanglement Module (DDM), which comprises Degradation Coupled Attention (DCA) to capture both general and weather-specific features, thereby enabling effective disentanglement and transfer of degradation patterns. Experimental results demonstrate that DDTNet significantly and consistently improves existing all-in-one models across real-world deraining, desnowing, and dehazing datasets.