Deep Unfolding Network for Image Super-Resolution

Learning-based single image super-resolution (SISR) methods are continuously showing superior effectiveness and efficiency over traditional model-based methods, largely due to the end-to-end training. However, different from model-based methods that can handle the SISR problem with different scale f...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 3214 - 3223
Hauptverfasser: Zhang, Kai, Van Gool, Luc, Timofte, Radu
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.01.2020
Schlagworte:
ISSN:1063-6919
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Learning-based single image super-resolution (SISR) methods are continuously showing superior effectiveness and efficiency over traditional model-based methods, largely due to the end-to-end training. However, different from model-based methods that can handle the SISR problem with different scale factors, blur kernels and noise levels under a unified MAP (maximum a posteriori) framework, learning-based methods generally lack such flexibility. To address this issue, this paper proposes an end-to-end trainable unfolding network which leverages both learningbased methods and model-based methods. Specifically, by unfolding the MAP inference via a half-quadratic splitting algorithm, a fixed number of iterations consisting of alternately solving a data subproblem and a prior subproblem can be obtained. The two subproblems then can be solved with neural modules, resulting in an end-to-end trainable, iterative network. As a result, the proposed network inherits the flexibility of model-based methods to super-resolve blurry, noisy images for different scale factors via a single model, while maintaining the advantages of learning-based methods. Extensive experiments demonstrate the superiority of the proposed deep unfolding network in terms of flexibility, effectiveness and also generalizability.
AbstractList Learning-based single image super-resolution (SISR) methods are continuously showing superior effectiveness and efficiency over traditional model-based methods, largely due to the end-to-end training. However, different from model-based methods that can handle the SISR problem with different scale factors, blur kernels and noise levels under a unified MAP (maximum a posteriori) framework, learning-based methods generally lack such flexibility. To address this issue, this paper proposes an end-to-end trainable unfolding network which leverages both learningbased methods and model-based methods. Specifically, by unfolding the MAP inference via a half-quadratic splitting algorithm, a fixed number of iterations consisting of alternately solving a data subproblem and a prior subproblem can be obtained. The two subproblems then can be solved with neural modules, resulting in an end-to-end trainable, iterative network. As a result, the proposed network inherits the flexibility of model-based methods to super-resolve blurry, noisy images for different scale factors via a single model, while maintaining the advantages of learning-based methods. Extensive experiments demonstrate the superiority of the proposed deep unfolding network in terms of flexibility, effectiveness and also generalizability.
Author Van Gool, Luc
Zhang, Kai
Timofte, Radu
Author_xml – sequence: 1
  givenname: Kai
  surname: Zhang
  fullname: Zhang, Kai
  organization: Computer Vision Lab, ETH Zurich, Switzerland
– sequence: 2
  givenname: Luc
  surname: Van Gool
  fullname: Van Gool, Luc
  organization: Computer Vision Lab, ETH Zurich, Switzerland
– sequence: 3
  givenname: Radu
  surname: Timofte
  fullname: Timofte, Radu
  organization: Computer Vision Lab, ETH Zurich, Switzerland
BookMark eNotjs1Og0AURkejibXyBLrgBcB7Z5hh7tLgX5NGTbVumwHuNChlCNAY394aXXw5m5OT71ycdKFjIa4QUkSg6-L9ZZVJA5BKkJACKGmPRES5xVwehsbqYzFDMCoxhHQmonH8gF8P0ZCdCbxl7uN150NbN902fuLpKwyfsQ9DvNi5Lcev-56HZMVjaPdTE7oLcepdO3L0z7lY39-9FY_J8vlhUdwsk0ahnhJrCTRRyTUow7aqmXTOKCk73C2lrcvKaZVXXjsoFRvjKsgs1CSBPfpMzcXlX7dh5k0_NDs3fG8IdQ4k1Q_FZUax
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR42600.2020.00328
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
EISBN 9781728171685
1728171687
EISSN 1063-6919
EndPage 3223
ExternalDocumentID 9157092
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i315t-8890599bed036e8cde957e1294600b28dbca537cf5a0b3e66ac0480d920ef1f43
IEDL.DBID RIE
ISICitedReferencesCount 551
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000620679503047&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:30:35 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i315t-8890599bed036e8cde957e1294600b28dbca537cf5a0b3e66ac0480d920ef1f43
OpenAccessLink http://hdl.handle.net/20.500.11850/460815
PageCount 10
ParticipantIDs ieee_primary_9157092
PublicationCentury 2000
PublicationDate 2020-01-01
PublicationDateYYYYMMDD 2020-01-01
PublicationDate_xml – month: 01
  year: 2020
  text: 2020-01-01
  day: 01
PublicationDecade 2020
PublicationTitle Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online)
PublicationTitleAbbrev CVPR
PublicationYear 2020
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0003211698
Score 2.6317265
Snippet Learning-based single image super-resolution (SISR) methods are continuously showing superior effectiveness and efficiency over traditional model-based...
SourceID ieee
SourceType Publisher
StartPage 3214
SubjectTerms Computational modeling
Degradation
Image resolution
Inference algorithms
Kernel
Learning systems
Noise level
Title Deep Unfolding Network for Image Super-Resolution
URI https://ieeexplore.ieee.org/document/9157092
WOSCitedRecordID wos000620679503047&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07TwMxDLZKxcBUoEW8lYGR0Nwld0nmQgVLVQFF3ao8fFIH2qoPfj9J7lQYWNiiLJGdxHYcf58B7rJgFZ2JiB7PBBXClVSz3FHvpWGVFB6FSc0m5GikplM9bsH9HguDiKn4DB_iMP3l-6XbxVRZX2eFZDoY3AMpyxqrtc-n8PCSKbVq0HEZ0_3Bx_g18a-HV2AeC7h4bLn-q4dKciHDzv8WP4beDxaPjPde5gRauDiFThM8kuZqbrqQPSKuyCScl_SdREZ1fTcJQSl5-QxWg7ztVrimMV9fn7YeTIZP74Nn2vRDoHOeFVuqlI5sKhZ9UDAq51EXEoPDFkFImytvnSm4dFVhmOVYlsZFxLjXOcMqqwQ_g_ZiucBzIEyFMMlHapvIEGgiI4w1VljutBe55RfQjRqYrWrKi1kj_OXf01dwFFVcZyauob1d7_AGDt3Xdr5Z36Z9-gZrxZJO
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED5VBQmmAi3ijQdGQp3YSey5ULWiRBW0qFvlx0Xq0If64PdjJ1FhYGGLvEQ-23fn833fB_AQOq9olEf0WMoDzk0SSBqZwNpU0TzlFrkqxCbSLBOTiRzW4HGPhUHEovkMn_xn8ZZvl2bnS2VtGcYplc7hHnjlrAqtta-oMHeXSaSo8HEhle3O5_C9YGB398DIt3AxL7r-S0WlCCLdxv9-fwKtHzQeGe7jzCnUcHEGjSp9JNXh3DQhfEZckbHbMcWDEsnKDm_i0lLSnzu_QT52K1wHvmJf7rcWjLsvo04vqBQRghkL420ghPR8KhqtMzEKY1HGKbqQzd0kdSSsNipmqcljRTXDJFHGY8atjCjmYc7ZOdQXywVeAKHCJUrWk9t4jkDlOWG00lwzIy2PNLuEprfAdFWSXkyryV_9PXwPR73R22A66Gev13DszV3WKW6gvl3v8BYOzdd2tlnfFWv2DT_klZc
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=Proceedings+%28IEEE+Computer+Society+Conference+on+Computer+Vision+and+Pattern+Recognition.+Online%29&rft.atitle=Deep+Unfolding+Network+for+Image+Super-Resolution&rft.au=Zhang%2C+Kai&rft.au=Van+Gool%2C+Luc&rft.au=Timofte%2C+Radu&rft.date=2020-01-01&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=3214&rft.epage=3223&rft_id=info:doi/10.1109%2FCVPR42600.2020.00328&rft.externalDocID=9157092