用预训练扩散模型做无损图像修复，无需知道退化类型。
Diffusion Image Prior
- 利用预训练扩散模型作为强先验，无需显式退化模型。
- 优化过程先恢复清晰图像，再过拟合到噪声输入，可覆盖多种退化。
- 通过早停实现盲修复，适用于去JPEG、去水渍、降噪、超分等任务。
基于预训练扩散模型的零样本图像修复方法近期取得显著进展，但通常需至少已知退化模型的形式。然而在真实场景中，退化可能过于复杂而无法明确描述。为此，我们提出扩散图像先验（DIIP）。受深度图像先验（DIP）启发，DIIP无需显式退化模型即可去除伪影。但与DIP不同，我们发现预训练扩散模型即使未在受损数据上训练，也提供更强先验。实验表明，DIIP的优化过程首先重建清晰图像，随后逐渐过拟合至退化输入，且适用退化范围更广。基于此，我们提出一种基于早停的盲图像修复方法，无需退化模型先验。在多种盲图像修复任务中验证，包括JPEG伪影消除、水滴去除、去噪和超分辨率，均达到当前最优效果。
原文摘要 · Abstract (English)
Zero-shot image restoration (IR) methods based on pretrained diffusion models have recently achieved significant success. These methods typically require at least a parametric form of the degradation model. However, in real-world scenarios, the degradation may be too complex to define explicitly. To handle this general case, we introduce the Diffusion Image Prior (DIIP). We take inspiration from the Deep Image Prior (DIP)[16], since it can be used to remove artifacts without the need for an explicit degradation model. However, in contrast to DIP, we find that pretrained diffusion models offer a much stronger prior, despite being trained without knowledge from corrupted data. We show that, the optimization process in DIIP first reconstructs a clean version of the image before eventually overfitting to the degraded input, but it does so for a broader range of degradations than DIP. In light of this result, we propose a blind image restoration (IR) method based on early stopping, which does not require prior knowledge of the degradation model. We validate DIIP on various degradation-blind IR tasks, including JPEG artifact removal, waterdrop removal, denoising and super-resolution with state-of-the-art results.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。