Perceive-IR: Learning to Perceive Degradation Better for All-in-One Image Restoration

TL;DR

Perceive-IR introduces a backbone-agnostic framework for fine-grained, multi-level image quality perception and restoration, enhancing adaptability and precision in handling various degradation types and severity levels.

Contribution

It proposes a modular, backbone-agnostic approach with a two-stage process for fine-grained quality perception and adaptive restoration, improving over existing methods.

Findings

Effective multi-level quality perception aligned with CLIP perception space

Seamless integration into advanced restoration models

Improved restoration quality through difficulty-adaptive perceptual loss

Abstract

Existing All-in-One image restoration methods often fail to perceive degradation types and severity levels simultaneously, overlooking the importance of fine-grained quality perception. Moreover, these methods often utilize highly customized backbones, which hinder their adaptability and integration into more advanced restoration networks. To address these limitations, we propose Perceive-IR, a novel backbone-agnostic All-in-One image restoration framework designed for fine-grained quality control across various degradation types and severity levels. Its modular structure allows core components to function independently of specific backbones, enabling seamless integration into advanced restoration models without significant modifications. Specifically, Perceive-IR operates in two key stages: 1) multi-level quality-driven prompt learning stage, where a fine-grained quality perceiver is meticulously trained to discern three tier quality levels by optimizing the alignment between prompts and images within the CLIP perception space. This stage ensures a nuanced understanding of image quality, laying the groundwork for subsequent restoration; 2) restoration stage, where the quality perceiver is seamlessly integrated with a difficulty-adaptive perceptual loss, forming a quality-aware learning strategy. This strategy not only dynamically differentiates sample learning difficulty but also achieves fine-grained quality control by driving the restored image toward the ground truth while pulling it away from both low- and medium-quality samples.