UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancement

TL;DR

UARE is the first unified vision-language model that jointly addresses image quality assessment, restoration, and enhancement, leveraging multi-task training to improve performance across these interconnected low-level vision tasks.

Contribution

The paper introduces UARE, a novel unified model that integrates IQA and restoration, enabling mutual guidance and improved results through a two-stage training framework.

Findings

UARE effectively handles multiple degradations.

Multi-task training boosts restoration and enhancement quality.

Extensive experiments validate UARE's superior performance.

Abstract

Image quality assessment (IQA) and image restoration are fundamental problems in low-level vision. Although IQA and restoration are closely connected conceptually, most existing work treats them in isolation. Recent advances in unified multimodal understanding-generation models demonstrate promising results and indicate that stronger understanding can improve generative performance. This motivates a single model that unifies IQA and restoration and explicitly studies how IQA can guide restoration, a setting that remains largely underexplored yet highly valuable. In this paper, we propose UARE, to our knowledge the first Unified vision-language model for image quality Assessment, Restoration, and Enhancement. Built on pretrained unified understanding and generation models, we introduce a two-stage training framework. First, a progressive, easy-to-hard schedule expands from single-type distortions to higher-order mixed degradations, enabling UARE to handle multiple degradations. Second, we perform unified fine-tuning of quality understanding and restoration with interleaved text-image data, aligning IQA signals with restoration objectives. Through multi-task co-training, UARE leverages IQA to boost restoration and enhancement performance. Extensive experiments across IQA, restoration, and enhancement tasks demonstrate the effectiveness of UARE. The code and models will be available at https://github.com/lwq20020127/UARE.