ConStyle v2：用于全能图像恢复的强提示器
ConStyle v2: A Strong Prompter for All-in-One Image Restoration
AI总结：
本文提出 ConStyle v2，一种即插即用提示器，通过两阶段训练和三项改进，能在不到两个 GPU 天内增强任意 U-Net 图像恢复模型以处理多种退化，且无需微调。
摘要：
本文介绍了 ConStyle v2，一个强大的即插即用提示器，旨在输出干净的视觉提示，并协助 U-Net 图像恢复模型处理多种退化。IRConStyle 是一个由 ConStyle 和一个通用恢复网络组成的图像恢复框架，其联合训练过程分为两个阶段：首先单独预训练 ConStyle，然后冻结其权重以指导通用恢复网络的训练。在预训练阶段提出了三项改进来训练 ConStyle：无监督预训练、添加前置任务（即分类）以及采用知识蒸馏。无需额外修饰，我们可以在不到两个 GPU 天内获得 ConStyle v2，一个用于全能图像恢复的强提示器，且无需任何微调。在 Restormer（基于 Transformer）、NAFNet（基于 CNN）、MAXIM-1S（基于 MLP）以及一个普通 CNN 网络上的大量实验表明，ConStyle v2 可以将任何 U-Net 风格的图像恢复模型增强为全能图像恢复模型。此外，由训练良好的 ConStyle v2 指导的模型在某些特定退化方面表现出优于 ConStyle 的性能。
英文摘要：
This paper introduces ConStyle v2, a strong plug-and-play prompter designed to output clean visual prompts and assist U-Net Image Restoration models in handling multiple degradations. The joint training process of IRConStyle, an Image Restoration framework consisting of ConStyle and a general restoration network, is divided into two stages: first, pre-training ConStyle alone, and then freezing its weights to guide the training of the general restoration network. Three improvements are proposed in the pre-training stage to train ConStyle: unsupervised pre-training, adding a pretext task (i.e. classification), and adopting knowledge distillation. Without bells and whistles, we can get ConStyle v2, a strong prompter for all-in-one Image Restoration, in less than two GPU days and doesn't require any fine-tuning. Extensive experiments on Restormer (transformer-based), NAFNet (CNN-based), MAXIM-1S (MLP-based), and a vanilla CNN network demonstrate that ConStyle v2 can enhance any U-Net style Image Restoration models to all-in-one Image Restoration models. Furthermore, models guided by the well-trained ConStyle v2 exhibit superior performance in some specific degradation compared to ConStyle.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
