复原、评估、重复：面向迭代图像复原的统一框架
Restore, Assess, Repeat: A Unified Framework for Iterative Image Restoration
把 IQA 作为复原流程中的评估环节嵌入潜空间，形成「复原—评估—再复原」的迭代闭环，可复用于处理未知或复合退化的 all-in-one 复原模型设计，并避免图像/文本解码带来的延迟与信息损失。
原文依据 we propose RAR, a Restore, Assess and Repeat process, that integrates Image Quality Assessment (IQA) and Image Restoration (IR) into a unified framework
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论图像复原旨在从受恶劣天气、模糊或低光照等多种因素退化的输入中恢复高质量图像。尽管近期研究在单项或统一复原任务上取得了显著进展，但在处理未知或复合退化时仍受限于泛化能力不足与效率低下。为解决这些局限，我们提出 RAR，即「复原、评估、重复」流程，它将图像质量评估（IQA）与图像复原（IR）整合进一个统一框架，以迭代且高效地实现高质量图像复原。具体而言，我们引入一个完全在潜变量域中运行的复原过程，联合执行退化识别、图像复原与质量验证。所得模型可完全端到端训练，并支持一种动态自适应复原过程的 all-in-one 评估与复原方法。此外，将 IQA 与 IR 紧密集成到单一模型中，可最大限度减少因两个模块彼此分离（例如在图像和/或文本解码期间）而通常产生的延迟与信息损失。大量实验表明，我们的方法在单一、未知与复合退化下均取得一致提升，从而确立了新的最先进水平。
英文原文摘要 1319 字符
Image restoration aims to recover high quality images from inputs degraded by various factors, such as adverse weather, blur, or low light. While recent studies have shown remarkable progress across individual or unified restoration tasks, they still suffer from limited generalization and inefficiency when handling unknown or composite degradations. To address these limitations, we propose RAR, a Restore, Assess and Repeat process, that integrates Image Quality Assessment (IQA) and Image Restoration (IR) into a unified framework to iteratively and efficiently achieve high quality image restoration. Specifically, we introduce a restoration process that operates entirely in the latent domain to jointly perform degradation identification, image restoration, and quality verification. The resulting model is fully trainable end to end and allows for an all-in-one assess and restore approach that dynamically adapts the restoration process. Also, the tight integration of IQA and IR into a unified model minimizes the latency and information loss that typically arises from keeping the two modules disjoint, (e.g. during image and/or text decoding). Extensive experiments show that our approach consistent improvements under single, unknown and composite degradations, thereby establishing a new state-of-the-art.