Dual Ascent Diffusion for Inverse Problems

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Kim, Minseo, Levy, Axel, Wetzstein, Gordon
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913124854005760
author Kim, Minseo
Levy, Axel
Wetzstein, Gordon
author_facet Kim, Minseo
Levy, Axel
Wetzstein, Gordon
contents Ill-posed inverse problems are fundamental in many domains, ranging from astrophysics to medical imaging. Emerging diffusion models provide a powerful prior for solving these problems. Existing maximum-a-posteriori (MAP) or posterior sampling approaches, however, rely on different computational approximations, leading to inaccurate or suboptimal samples. To address this issue, we introduce a new approach to solving MAP problems with diffusion model priors using a dual ascent optimization framework. Our framework achieves better image quality as measured by various metrics for image restoration problems, it is more robust to high levels of measurement noise, it is faster, and it estimates solutions that represent the observations more faithfully than the state of the art.
format Preprint
id arxiv_https___arxiv_org_abs_2505_17353
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dual Ascent Diffusion for Inverse Problems
Kim, Minseo
Levy, Axel
Wetzstein, Gordon
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Image and Video Processing
Ill-posed inverse problems are fundamental in many domains, ranging from astrophysics to medical imaging. Emerging diffusion models provide a powerful prior for solving these problems. Existing maximum-a-posteriori (MAP) or posterior sampling approaches, however, rely on different computational approximations, leading to inaccurate or suboptimal samples. To address this issue, we introduce a new approach to solving MAP problems with diffusion model priors using a dual ascent optimization framework. Our framework achieves better image quality as measured by various metrics for image restoration problems, it is more robust to high levels of measurement noise, it is faster, and it estimates solutions that represent the observations more faithfully than the state of the art.
title Dual Ascent Diffusion for Inverse Problems
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Image and Video Processing
url https://arxiv.org/abs/2505.17353