ClearAIR: A Human-Visual-Perception-Inspired All-in-One Image Restoration

TL;DR

ClearAIR tackles All-in-One Image Restoration by modeling human visual perception in a coarse-to-fine pipeline. It fuses global quality assessment from an MLLM-IQA, region-aware semantic guidance, degradation-aware local adaptation, and self-supervised fine-detail refinement through an Internal Clue Reuse Mechanism, organized in a global-to-local sequence with a unifying loss $L_{total}=L_1+\alpha\,L_{inter}$ and $\alpha=0.25$. Empirical results on synthetic and real-world datasets show state-of-the-art performance across three- and five-degradation scenarios, All-Weather conditions, and composite degradations, with improved texture fidelity and perceptual quality. The work demonstrates practical impact for robust restoration in real-world settings and points to perceptual-adaptive extensions, such as integrating a Just Noticeable Difference (JND) mechanism for dynamic routing and perceptual weighting.

Abstract

All-in-One Image Restoration (AiOIR) has advanced significantly, offering promising solutions for complex real-world degradations. However, most existing approaches rely heavily on degradation-specific representations, often resulting in oversmoothing and artifacts. To address this, we propose ClearAIR, a novel AiOIR framework inspired by Human Visual Perception (HVP) and designed with a hierarchical, coarse-to-fine restoration strategy. First, leveraging the global priority of early HVP, we employ a Multimodal Large Language Model (MLLM)-based Image Quality Assessment (IQA) model for overall evaluation. Unlike conventional IQA, our method integrates cross-modal understanding to more accurately characterize complex, composite degradations. Building upon this overall assessment, we then introduce a region awareness and task recognition pipeline. A semantic cross-attention, leveraging semantic guidance unit, first produces coarse semantic prompts. Guided by this regional context, a degradation-aware module implicitly captures region-specific degradation characteristics, enabling more precise local restoration. Finally, to recover fine details, we propose an internal clue reuse mechanism. It operates in a self-supervised manner to mine and leverage the intrinsic information of the image itself, substantially enhancing detail restoration. Experimental results show that ClearAIR achieves superior performance across diverse synthetic and real-world datasets.

Figures (9)

• Figure 1: A coarse-to-fine image processing pipeline inspired by human visual perception.
• Figure 2: An overview architecture of the proposed ClearAIR.
• Figure 3: The proposed Internal Clue Reuse Mechanism.
• Figure 4: Visual comparisons of ClearAIR with state-of-the-art AiOIR methods on Three Degradations task.
• Figure 5: Visual comparisons of ClearAIR with state-of-the-art AiOIR methods on All-Weather task.