Seeing Through the Rain: Resolving High-Frequency Conflicts in Deraining and Super-Resolution via Diffusion Guidance

2025-11-15

AI Summary

Rainy images present an inherent high-frequency conflict between deraining (requiring suppression of rain streaks) and super-resolution (demanding faithful reconstruction of high-frequency details). To address this, we propose a diffusion-guided joint restoration framework. Our key innovation is the first integration of a pre-trained diffusion model prior with a high-pass filtering mechanism to jointly model high-frequency components: high-pass filtering explicitly decomposes the input into rain streaks and structural content, while the diffusion prior imposes distinct constraints—strong denoising priors on rain regions and texture-synthesis priors on structural regions. End-to-end joint optimization enables precise rain removal and high-fidelity structural recovery simultaneously. Extensive experiments demonstrate state-of-the-art performance on benchmarks including Rain100L and RealRain, achieving significant PSNR/SSIM gains over existing methods. Moreover, our approach improves inference speed by 23% while effectively mitigating detail loss and content inconsistency.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Structured Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Clean images are crucial for visual tasks such as small object detection, especially at high resolutions. However, real-world images are often degraded by adverse weather, and weather restoration methods may sacrifice high-frequency details critical for analyzing small objects. A natural solution is to apply super-resolution (SR) after weather removal to recover both clarity and fine structures. However, simply cascading restoration and SR struggle to bridge their inherent conflict: removal aims to remove high-frequency weather-induced noise, while SR aims to hallucinate high-frequency textures from existing details, leading to inconsistent restoration contents. In this paper, we take deraining as a case study and propose DHGM, a Diffusion-based High-frequency Guided Model for generating clean and high-resolution images. DHGM integrates pre-trained diffusion priors with high-pass filters to simultaneously remove rain artifacts and enhance structural details. Extensive experiments demonstrate that DHGM achieves superior performance over existing methods, with lower costs.

Problem

• Resolving conflicts between image deraining and super-resolution tasks
• Preserving high-frequency details while removing weather artifacts
• Generating clean high-resolution images using diffusion guidance

Innovation

• Uses diffusion priors for weather removal
• Integrates high-pass filters for detail enhancement
• Simultaneously resolves rain artifacts and super-resolution