DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Hengkang, Zhang, Xu, Li, Taihui, Wan, Yuxiang, Chen, Tiancong, Sun, Ju
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929578292805632
author Wang, Hengkang
Zhang, Xu
Li, Taihui
Wan, Yuxiang
Chen, Tiancong
Sun, Ju
author_facet Wang, Hengkang
Zhang, Xu
Li, Taihui
Wan, Yuxiang
Chen, Tiancong
Sun, Ju
contents Pretrained diffusion models (DMs) have recently been popularly used in solving inverse problems (IPs). The existing methods mostly interleave iterative steps in the reverse diffusion process and iterative steps to bring the iterates closer to satisfying the measurement constraint. However, such interleaving methods struggle to produce final results that look like natural objects of interest (i.e., manifold feasibility) and fit the measurement (i.e., measurement feasibility), especially for nonlinear IPs. Moreover, their capabilities to deal with noisy IPs with unknown types and levels of measurement noise are unknown. In this paper, we advocate viewing the reverse process in DMs as a function and propose a novel plug-in method for solving IPs using pretrained DMs, dubbed DMPlug. DMPlug addresses the issues of manifold feasibility and measurement feasibility in a principled manner, and also shows great potential for being robust to unknown types and levels of noise. Through extensive experiments across various IP tasks, including two linear and three nonlinear IPs, we demonstrate that DMPlug consistently outperforms state-of-the-art methods, often by large margins especially for nonlinear IPs. The code is available at https://github.com/sun-umn/DMPlug.
format Preprint
id arxiv_https___arxiv_org_abs_2405_16749
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Models
Wang, Hengkang
Zhang, Xu
Li, Taihui
Wan, Yuxiang
Chen, Tiancong
Sun, Ju
Machine Learning
Computer Vision and Pattern Recognition
Pretrained diffusion models (DMs) have recently been popularly used in solving inverse problems (IPs). The existing methods mostly interleave iterative steps in the reverse diffusion process and iterative steps to bring the iterates closer to satisfying the measurement constraint. However, such interleaving methods struggle to produce final results that look like natural objects of interest (i.e., manifold feasibility) and fit the measurement (i.e., measurement feasibility), especially for nonlinear IPs. Moreover, their capabilities to deal with noisy IPs with unknown types and levels of measurement noise are unknown. In this paper, we advocate viewing the reverse process in DMs as a function and propose a novel plug-in method for solving IPs using pretrained DMs, dubbed DMPlug. DMPlug addresses the issues of manifold feasibility and measurement feasibility in a principled manner, and also shows great potential for being robust to unknown types and levels of noise. Through extensive experiments across various IP tasks, including two linear and three nonlinear IPs, we demonstrate that DMPlug consistently outperforms state-of-the-art methods, often by large margins especially for nonlinear IPs. The code is available at https://github.com/sun-umn/DMPlug.
title DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Models
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.16749