Diffusion Posterior Proximal Sampling for Image Restoration

arXiv:2402.16907v1 Announce Type: cross
Abstract: Diffusion models have demonstrated remarkable efficacy in generating high-quality samples. Existing diffusion-based image restoration algorithms exploit pre-trained diffusion models to leverage data priors, yet they still preserve elements inherited from the unconditional generation paradigm. These strategies initiate the denoising process with pure white noise and incorporate random noise at each generative step, leading to over-smoothed results. In this paper, we introduce a refined paradigm for diffusion-based image restoration. Specifically, we opt for a sample consistent …

abstract algorithms arxiv cs.cv cs.lg data denoising diffusion diffusion models eess.iv exploit image image restoration noise paradigm posterior process quality random samples sampling strategies type white noise