Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration

TL;DR

This paper introduces a novel domain adaptation method for image restoration that leverages diffusion models in the noise space, effectively aligning synthetic and real-world data distributions to improve restoration performance.

Contribution

It proposes a new diffusion-based domain adaptation technique called denoising as adaptation, with strategies to prevent shortcuts and enhance stability in low-level vision tasks.

Findings

Effective in denoising, deblurring, and deraining tasks

Outperforms existing domain adaptation methods

Improves generalization to real-world scenarios

Abstract

Although learning-based image restoration methods have made significant progress, they still struggle with limited generalization to real-world scenarios due to the substantial domain gap caused by training on synthetic data. Existing methods address this issue by improving data synthesis pipelines, estimating degradation kernels, employing deep internal learning, and performing domain adaptation and regularization. Previous domain adaptation methods have sought to bridge the domain gap by learning domain-invariant knowledge in either feature or pixel space. However, these techniques often struggle to extend to low-level vision tasks within a stable and compact framework. In this paper, we show that it is possible to perform domain adaptation via the noise space using diffusion models. In particular, by leveraging the unique property of how auxiliary conditional inputs influence the multi-step denoising process, we derive a meaningful diffusion loss that guides the restoration model in progressively aligning both restored synthetic and real-world outputs with a target clean distribution. We refer to this method as denoising as adaptation. To prevent shortcuts during joint training, we present crucial strategies such as channel-shuffling layer and residual-swapping contrastive learning in the diffusion model. They implicitly blur the boundaries between conditioned synthetic and real data and prevent the reliance of the model on easily distinguishable features. Experimental results on three classical image restoration tasks, namely denoising, deblurring, and deraining, demonstrate the effectiveness of the proposed method.