Dynamic Degradation Decomposition Network for All-in-One Image Restoration

2025-02-26

AI Summary

Addressing the challenges of unified restoration and poor generalization across diverse image degradation types, this paper proposes the Dynamic Degradation Decomposition Network (D³Net). Methodologically, we introduce the first Cross-Domain Degradation Analyzer (CDDA) to identify unknown degradation categories; further, we design a prompt-driven Dynamic Decomposition Mechanism (DDM) that integrates frequency-domain modeling, spatial feature interaction, and dual-level prompting for degradation-adaptive restoration. The core technical innovations are the two-level prompting strategy and progressive degradation decomposition, which jointly balance modeling accuracy and computational efficiency. On the SOTS-Outdoor and GoPro datasets, D³Net achieves PSNR improvements of 5.47 dB and 3.30 dB, respectively, significantly outperforming state-of-the-art methods. These results demonstrate D³Net’s strong generalization capability and robustness under complex, heterogeneous degradation scenarios.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Currently, restoring clean images from a variety of degradation types using a single model is still a challenging task. Existing all-in-one image restoration approaches struggle with addressing complex and ambiguously defined degradation types. In this paper, we introduce a dynamic degradation decomposition network for all-in-one image restoration, named D$^3$Net. D$^3$Net achieves degradation-adaptive image restoration with guided prompt through cross-domain interaction and dynamic degradation decomposition. Concretely, in D$^3$Net, the proposed Cross-Domain Degradation Analyzer (CDDA) engages in deep interaction between frequency domain degradation characteristics and spatial domain image features to identify and model variations of different degradation types on the image manifold, generating degradation correction prompt and strategy prompt, which guide the following decomposition process. Furthermore, the prompt-based Dynamic Decomposition Mechanism (DDM) for progressive degradation decomposition, that encourages the network to adaptively select restoration strategies utilizing the two-level prompt generated by CDDA. Thanks to the synergistic cooperation between CDDA and DDM, D$^3$Net achieves superior flexibility and scalability in handling unknown degradation, while effectively reducing unnecessary computational overhead. Extensive experiments on multiple image restoration tasks demonstrate that D$^3$Net significantly outperforms the state-of-the-art approaches, especially improving PSNR by 5.47dB and 3.30dB on the SOTS-Outdoor and GoPro datasets, respectively.

Problem

• Single model for diverse image degradation
• Dynamic degradation decomposition network
• Cross-domain interaction for image restoration

Innovation

• Dynamic degradation decomposition network
• Cross-domain interaction analysis
• Prompt-based adaptive restoration