Degradation-Aware Adaptive Context Gating for Unified Image Restoration

2026-05-02

AI Summary

This work addresses the challenge of task interference arising from coexisting multiple degradations in unified image restoration by proposing DACG-IR, a novel method that leverages a lightweight multi-scale degradation-aware module to generate inter-layer prompts. These prompts, combined with degradation-aware contextual modeling, dynamically modulate attention temperature and output gating to enhance restoration fidelity. Furthermore, the method introduces a spatial-channel dual-gating adaptive fusion mechanism that effectively suppresses the propagation of shallow-layer noise into deeper layers. Extensive experiments demonstrate that DACG-IR consistently outperforms state-of-the-art approaches across diverse settings, including single-task restoration, unified all-in-one restoration, adverse weather removal, and complex composite degradation scenarios.

Abstract

Unified image restoration using a single model often faces task interference due to diverse degradations. To address this, we propose DACG-IR (Degradation-Aware Adaptive Context Gating), which enables explicit perception of degradation characteristics to dynamically modulate feature representations. Our method constructs degradation-aware contextual representations from the input to modulate attention distribution, frequency-domain features, and feature aggregation. Specifically, a lightweight multi-scale degradation-aware module extracts coarse degradation information and generates layer-wise prompts. These prompts guide attention temperature and output gating in encoder and decoder blocks for adaptive feature extraction. Additionally, a spatial-channel dual-gated adaptive fusion mechanism refines encoder features, suppressing noise propagation from shallow to deep layers. This design effectively suppresses degradation-induced noise while preserving informative structures. Experiments show DACG-IR outperforms state-of-the-art methods in single-task, all-in-one, adverse weather removal, and composite degradation settings. Code: https://github.com/HlHomes/DACG-IR-code

Problem

• unified image restoration
• task interference
• degradation diversity
• image degradation
• single-model restoration

Innovation

• Degradation-aware
• Adaptive Context Gating
• Unified Image Restoration
• Dynamic Feature Modulation
• Dual-gated Fusion