面向一体化图像复原的退化一致性测试时自适应
Degradation-Consistent Test-Time Adaptation for All-in-One Image Restoration
把「同一场景的多种退化版本应映射到同一干净图像」作为自监督一致性约束，配合扩散生成器合成伪退化-干净对，为图像复原模型在测试时做无源域自适应提供了可复用的伪监督构造思路；其按退化敏感性选择参数更新的策略也可迁移到其他复原/增强模型的稳定微调。
原文依据 multiple degraded versions of the same scene should map to a consistent clean image
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论一体化图像复原（AiOIR）方法在处理多种退化方面已取得显著进展。然而，当测试分布偏离训练分布时，其性能往往下降。因此，探索面向 AiOIR 的测试时自适应至关重要。为在无法访问源数据、也不重新训练的情况下，使预训练的 AiOIR 模型适应未见过的退化分布，必须解决两个关键挑战：设计可靠的伪监督以及稳定自适应过程。观察到同一场景的多个退化版本应映射到一致的干净图像，我们提出退化一致性测试时自适应（DCTTA）。DCTTA 包含三个核心组件：（1）测试时再退化生成，利用基于扩散的生成器构造伪退化-干净图像对以实现分布对齐；（2）退化引导的图像复原，通过自监督一致性损失施加域自适应；（3）测试时重要参数选择，有选择地更新对退化敏感的参数，在保留预训练知识的同时确保自适应稳定。在多个任务和具有挑战性的域偏移上的大量实验表明，DCTTA 持续优于最优的 AiOIR 基线，在 Rain100H 数据集上取得最高 +4.57 dB 的峰值信噪比提升。
英文原文摘要 1373 字符
All-in-one image restoration (AiOIR) methods have made remarkable progress in handling diverse degradations. However, their performance often deteriorates when the test distribution deviates from the training distribution. Exploring test-time adaptation for AiOIR is therefore crucial. To adapt a pre-trained AiOIR model to unseen degradation distributions without access to source data or retraining, two key challenges must be addressed: designing reliable pseudo-supervision and stabilizing adaptation. Observing that multiple degraded versions of the same scene should map to a consistent clean image, we propose Degradation-Consistent Test-Time Adaptation (DCTTA). DCTTA comprises three core components: (1) test-time redegradation generation, which leverages a diffusion-based generator to construct pseudo degraded-clean pairs for distribution alignment; (2) degradation-guided image restoration, which enforces domain adaptation via self-supervised consistency loss; and (3) test-time important parameter selection, which selectively updates degradation-sensitive parameters to ensure stable adaptation while preserving pre-trained knowledge. Extensive experiments across multiple tasks and challenging domain shifts demonstrate that DCTTA consistently outperforms state-of-the-art AiOIR baselines, achieving up to +4.57 dB PSNR improvement on the Rain100H dataset.