Equivariant Denoisers for Image Restoration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Renaud, Marien, Leclaire, Arthur, Papadakis, Nicolas
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910849341325312
author Renaud, Marien
Leclaire, Arthur
Papadakis, Nicolas
author_facet Renaud, Marien
Leclaire, Arthur
Papadakis, Nicolas
contents One key ingredient of image restoration is to define a realistic prior on clean images to complete the missing information in the observation. State-of-the-art restoration methods rely on a neural network to encode this prior. Moreover, typical image distributions are invariant to some set of transformations, such as rotations or flips. However, most deep architectures are not designed to represent an invariant image distribution. Recent works have proposed to overcome this difficulty by including equivariance properties within a Plug-and-Play paradigm. In this work, we propose a unified framework named Equivariant Regularization by Denoising (ERED) based on equivariant denoisers and stochastic optimization. We analyze the convergence of this algorithm and discuss its practical benefit.
format Preprint
id arxiv_https___arxiv_org_abs_2412_05343
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Equivariant Denoisers for Image Restoration
Renaud, Marien
Leclaire, Arthur
Papadakis, Nicolas
Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
One key ingredient of image restoration is to define a realistic prior on clean images to complete the missing information in the observation. State-of-the-art restoration methods rely on a neural network to encode this prior. Moreover, typical image distributions are invariant to some set of transformations, such as rotations or flips. However, most deep architectures are not designed to represent an invariant image distribution. Recent works have proposed to overcome this difficulty by including equivariance properties within a Plug-and-Play paradigm. In this work, we propose a unified framework named Equivariant Regularization by Denoising (ERED) based on equivariant denoisers and stochastic optimization. We analyze the convergence of this algorithm and discuss its practical benefit.
title Equivariant Denoisers for Image Restoration
topic Image and Video Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2412.05343