Cross-Scenario Deraining Adaptation with Unpaired Data: Superpixel Structural Priors and Multi-Stage Pseudo-Rain Synthesis

2026-03-23

AI Summary

This work addresses the significant performance degradation of existing image deraining methods in out-of-distribution (OOD) real-world scenarios, primarily caused by domain gaps between synthetic training data and authentic rainy conditions. To bridge this gap, we propose the first cross-scene deraining adaptation framework that requires only rain-free background images from the target domain—eliminating the need for paired rainy/clean image data. Our approach leverages a superpixel-based structural prior (via SLIC) to align structural information across domains and integrates a multi-stage pseudo-rain synthesis strategy with an iterative pseudo-label refinement mechanism to enable effective domain adaptation. The framework operates as a plug-and-play module compatible with any deraining model, yielding substantial improvements: it boosts PSNR by 32%–59% for state-of-the-art models under OOD conditions and markedly accelerates training convergence.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Image deraining plays a pivotal role in low-level computer vision, serving as a prerequisite for robust outdoor surveillance and autonomous driving systems. While deep learning paradigms have achieved remarkable success in firmly aligned settings, they often suffer from severe performance degradation when generalized to unseen Out-of-Distribution (OOD) scenarios. This failure stems primarily from the significant domain discrepancy between synthetic training datasets and the complex physical dynamics of real-world rain. To address these challenges, this paper proposes a pioneering cross-scenario deraining adaptation framework. Diverging from conventional approaches, our method obviates the requirements for paired rainy observations in the target domain, leveraging exclusively rain-free background images. We design a Superpixel Generation (Sup-Gen) module to extract stable structural priors from the source domain using Simple Linear Iterative Clustering. Subsequently, a Resolution-adaptive Fusion strategy is introduced to align these source structures with target backgrounds through texture similarity, ensuring the synthesis of diverse and realistic pseudo-data. Finally, we implement a pseudo-label re-Synthesize mechanism that employs multi-stage noise generation to simulate realistic rain streaks. This framework functions as a versatile plug-and-play module capable of seamless integration into arbitrary deraining architectures. Extensive experiments on state-of-the-art models demonstrate that our approach yields remarkable PSNR gains of up to 32% to 59% in OOD domains while significantly accelerating training convergence.

Problem

• cross-scenario deraining
• domain discrepancy
• unpaired data
• out-of-distribution generalization
• image deraining

Innovation

• cross-scenario adaptation
• unpaired data
• superpixel structural priors
• pseudo-rain synthesis
• multi-stage noise generation