Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution

Using diffusion models to solve inverse problems is a growing field of research. Current methods assume the degradation to be known and provide impressive results in terms of restoration quality and diversity. In this work, we leverage the efficiency of those models to jointly estimate the restored...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-11
Hauptverfasser: Laroche, Charles, Almansa, Andrés, Coupete, Eva
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Laroche, Charles
Almansa, Andrés
Coupete, Eva
description Using diffusion models to solve inverse problems is a growing field of research. Current methods assume the degradation to be known and provide impressive results in terms of restoration quality and diversity. In this work, we leverage the efficiency of those models to jointly estimate the restored image and unknown parameters of the degradation model such as blur kernel. In particular, we designed an algorithm based on the well-known Expectation-Minimization (EM) estimation method and diffusion models. Our method alternates between approximating the expected log-likelihood of the inverse problem using samples drawn from a diffusion model and a maximization step to estimate unknown model parameters. For the maximization step, we also introduce a novel blur kernel regularization based on a Plug \& Play denoiser. Diffusion models are long to run, thus we provide a fast version of our algorithm. Extensive experiments on blind image deblurring demonstrate the effectiveness of our method when compared to other state-of-the-art approaches.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2860454518</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2860454518</sourcerecordid><originalsourceid>FETCH-proquest_journals_28604545183</originalsourceid><addsrcrecordid>eNqNjEELgjAcxUcQJOV3-ENnQacz6VpKl27dZbqNJnMz_9O-fgrRudPjvd97b0MCmqZJVGSU7kiI2MVxTPMTZSwNCK84erhqpSbUzkJ5PwMH8fO9E9KAciM0RlsB2s5yRAnD6Boje4S39k_gw2B0y_268A6EbJ2dnZnW4EC2ihuU4Vf35FiVj8stWi5ek0Rfd24a7YJqWuRxxjKWFOl_rQ8mbEWC</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2860454518</pqid></control><display><type>article</type><title>Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution</title><source>Free E- Journals</source><creator>Laroche, Charles ; Almansa, Andrés ; Coupete, Eva</creator><creatorcontrib>Laroche, Charles ; Almansa, Andrés ; Coupete, Eva</creatorcontrib><description>Using diffusion models to solve inverse problems is a growing field of research. Current methods assume the degradation to be known and provide impressive results in terms of restoration quality and diversity. In this work, we leverage the efficiency of those models to jointly estimate the restored image and unknown parameters of the degradation model such as blur kernel. In particular, we designed an algorithm based on the well-known Expectation-Minimization (EM) estimation method and diffusion models. Our method alternates between approximating the expected log-likelihood of the inverse problem using samples drawn from a diffusion model and a maximization step to estimate unknown model parameters. For the maximization step, we also introduce a novel blur kernel regularization based on a Plug \&amp; Play denoiser. Diffusion models are long to run, thus we provide a fast version of our algorithm. Extensive experiments on blind image deblurring demonstrate the effectiveness of our method when compared to other state-of-the-art approaches.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Degradation ; Diffusion rate ; Image restoration ; Inverse problems ; Mathematical models ; Maximization ; Optimization ; Parameters ; Regularization</subject><ispartof>arXiv.org, 2023-11</ispartof><rights>2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>780,784</link.rule.ids></links><search><creatorcontrib>Laroche, Charles</creatorcontrib><creatorcontrib>Almansa, Andrés</creatorcontrib><creatorcontrib>Coupete, Eva</creatorcontrib><title>Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution</title><title>arXiv.org</title><description>Using diffusion models to solve inverse problems is a growing field of research. Current methods assume the degradation to be known and provide impressive results in terms of restoration quality and diversity. In this work, we leverage the efficiency of those models to jointly estimate the restored image and unknown parameters of the degradation model such as blur kernel. In particular, we designed an algorithm based on the well-known Expectation-Minimization (EM) estimation method and diffusion models. Our method alternates between approximating the expected log-likelihood of the inverse problem using samples drawn from a diffusion model and a maximization step to estimate unknown model parameters. For the maximization step, we also introduce a novel blur kernel regularization based on a Plug \&amp; Play denoiser. Diffusion models are long to run, thus we provide a fast version of our algorithm. Extensive experiments on blind image deblurring demonstrate the effectiveness of our method when compared to other state-of-the-art approaches.</description><subject>Algorithms</subject><subject>Degradation</subject><subject>Diffusion rate</subject><subject>Image restoration</subject><subject>Inverse problems</subject><subject>Mathematical models</subject><subject>Maximization</subject><subject>Optimization</subject><subject>Parameters</subject><subject>Regularization</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNjEELgjAcxUcQJOV3-ENnQacz6VpKl27dZbqNJnMz_9O-fgrRudPjvd97b0MCmqZJVGSU7kiI2MVxTPMTZSwNCK84erhqpSbUzkJ5PwMH8fO9E9KAciM0RlsB2s5yRAnD6Boje4S39k_gw2B0y_268A6EbJ2dnZnW4EC2ihuU4Vf35FiVj8stWi5ek0Rfd24a7YJqWuRxxjKWFOl_rQ8mbEWC</recordid><startdate>20231106</startdate><enddate>20231106</enddate><creator>Laroche, Charles</creator><creator>Almansa, Andrés</creator><creator>Coupete, Eva</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20231106</creationdate><title>Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution</title><author>Laroche, Charles ; Almansa, Andrés ; Coupete, Eva</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28604545183</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Algorithms</topic><topic>Degradation</topic><topic>Diffusion rate</topic><topic>Image restoration</topic><topic>Inverse problems</topic><topic>Mathematical models</topic><topic>Maximization</topic><topic>Optimization</topic><topic>Parameters</topic><topic>Regularization</topic><toplevel>online_resources</toplevel><creatorcontrib>Laroche, Charles</creatorcontrib><creatorcontrib>Almansa, Andrés</creatorcontrib><creatorcontrib>Coupete, Eva</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Laroche, Charles</au><au>Almansa, Andrés</au><au>Coupete, Eva</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution</atitle><jtitle>arXiv.org</jtitle><date>2023-11-06</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Using diffusion models to solve inverse problems is a growing field of research. Current methods assume the degradation to be known and provide impressive results in terms of restoration quality and diversity. In this work, we leverage the efficiency of those models to jointly estimate the restored image and unknown parameters of the degradation model such as blur kernel. In particular, we designed an algorithm based on the well-known Expectation-Minimization (EM) estimation method and diffusion models. Our method alternates between approximating the expected log-likelihood of the inverse problem using samples drawn from a diffusion model and a maximization step to estimate unknown model parameters. For the maximization step, we also introduce a novel blur kernel regularization based on a Plug \&amp; Play denoiser. Diffusion models are long to run, thus we provide a fast version of our algorithm. Extensive experiments on blind image deblurring demonstrate the effectiveness of our method when compared to other state-of-the-art approaches.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-11
issn 2331-8422
language eng
recordid cdi_proquest_journals_2860454518
source Free E- Journals
subjects Algorithms
Degradation
Diffusion rate
Image restoration
Inverse problems
Mathematical models
Maximization
Optimization
Parameters
Regularization
title Fast Diffusion EM: a diffusion model for blind inverse problems with application to deconvolution
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T08%3A50%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Fast%20Diffusion%20EM:%20a%20diffusion%20model%20for%20blind%20inverse%20problems%20with%20application%20to%20deconvolution&rft.jtitle=arXiv.org&rft.au=Laroche,%20Charles&rft.date=2023-11-06&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2860454518%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2860454518&rft_id=info:pmid/&rfr_iscdi=true