Saved in:
Bibliographic Details
Main Authors: Jayaram, Vivek, Kemelmacher-Shlizerman, Ira, Seitz, Steven M., Thickstun, John
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.00359
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909929330180096
author Jayaram, Vivek
Kemelmacher-Shlizerman, Ira
Seitz, Steven M.
Thickstun, John
author_facet Jayaram, Vivek
Kemelmacher-Shlizerman, Ira
Seitz, Steven M.
Thickstun, John
contents We introduce Linearly Constrained Diffusion Implicit Models (CDIM), a fast and accurate approach to solving noisy linear inverse problems using diffusion models. Traditional diffusion-based inverse methods rely on numerous projection steps to enforce measurement consistency in addition to unconditional denoising steps. CDIM achieves a 10-50x reduction in projection steps by dynamically adjusting the number and size of projection steps to align a residual measurement energy with its theoretical distribution under the forward diffusion process. This adaptive alignment preserves measurement consistency while substantially accelerating constrained inference. For noise-free linear inverse problems, CDIM exactly satisfies the measurement constraints with few projection steps, even when existing methods fail. We demonstrate CDIM's effectiveness across a range of applications, including super-resolution, denoising, inpainting, deblurring, and 3D point cloud reprojection. Code and an interactive demo can be found on our project website.
format Preprint
id arxiv_https___arxiv_org_abs_2411_00359
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Linearly Constrained Diffusion Implicit Models
Jayaram, Vivek
Kemelmacher-Shlizerman, Ira
Seitz, Steven M.
Thickstun, John
Machine Learning
Image and Video Processing
We introduce Linearly Constrained Diffusion Implicit Models (CDIM), a fast and accurate approach to solving noisy linear inverse problems using diffusion models. Traditional diffusion-based inverse methods rely on numerous projection steps to enforce measurement consistency in addition to unconditional denoising steps. CDIM achieves a 10-50x reduction in projection steps by dynamically adjusting the number and size of projection steps to align a residual measurement energy with its theoretical distribution under the forward diffusion process. This adaptive alignment preserves measurement consistency while substantially accelerating constrained inference. For noise-free linear inverse problems, CDIM exactly satisfies the measurement constraints with few projection steps, even when existing methods fail. We demonstrate CDIM's effectiveness across a range of applications, including super-resolution, denoising, inpainting, deblurring, and 3D point cloud reprojection. Code and an interactive demo can be found on our project website.
title Linearly Constrained Diffusion Implicit Models
topic Machine Learning
Image and Video Processing
url https://arxiv.org/abs/2411.00359