SUD$^2$: Supervision by Denoising Diffusion Models for Image Reconstruction
This addresses a challenge in imaging inverse problems like in-painting and dehazing for researchers and practitioners, though it appears incremental as it builds on existing denoising methods.
The paper tackles the problem of training image reconstruction networks when paired training data is scarce, by proposing a framework that uses denoising diffusion models to supervise network training, achieving results without the need for vast paired datasets.
Many imaging inverse problems$\unicode{x2014}$such as image-dependent in-painting and dehazing$\unicode{x2014}$are challenging because their forward models are unknown or depend on unknown latent parameters. While one can solve such problems by training a neural network with vast quantities of paired training data, such paired training data is often unavailable. In this paper, we propose a generalized framework for training image reconstruction networks when paired training data is scarce. In particular, we demonstrate the ability of image denoising algorithms and, by extension, denoising diffusion models to supervise network training in the absence of paired training data.