Learning Image Priors through Patch-based Diffusion Models for Solving Inverse Problems

TL;DR

This paper introduces PaDIS, a patch-based diffusion model that efficiently learns image priors for high-resolution inverse problems, outperforming previous methods especially with limited training data.

Contribution

The paper presents a novel patch-based, position-aware diffusion model that improves memory and data efficiency for solving high-dimensional inverse problems.

Findings

PaDIS achieves better memory and data efficiency.

It effectively solves inverse problems like CT reconstruction, deblurring, and superresolution.

Outperforms prior methods with limited training data.

Abstract

Diffusion models can learn strong image priors from underlying data distribution and use them to solve inverse problems, but the training process is computationally expensive and requires lots of data. Such bottlenecks prevent most existing works from being feasible for high-dimensional and high-resolution data such as 3D images. This paper proposes a method to learn an efficient data prior for the entire image by training diffusion models only on patches of images. Specifically, we propose a patch-based position-aware diffusion inverse solver, called PaDIS, where we obtain the score function of the whole image through scores of patches and their positional encoding and utilize this as the prior for solving inverse problems. First of all, we show that this diffusion model achieves an improved memory efficiency and data efficiency while still maintaining the capability to generate entire images via positional encoding. Additionally, the proposed PaDIS model is highly flexible and can be plugged in with different diffusion inverse solvers (DIS). We demonstrate that the proposed PaDIS approach enables solving various inverse problems in both natural and medical image domains, including CT reconstruction, deblurring, and superresolution, given only patch-based priors. Notably, PaDIS outperforms previous DIS methods trained on entire image priors in the case of limited training data, demonstrating the data efficiency of our proposed approach by learning patch-based prior.