Unsupervised Real Image Super-Resolution via Generative Variational AutoEncoder

Benefited from the deep learning, image Super-Resolution has been one of the most developing research fields in computer vision. Depending upon whether using a discriminator or not, a deep convolutional neural network can provide an image with high fidelity or better perceptual quality. Due to the l...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE Computer Society Conference on Computer Vision and Pattern Recognition workshops s. 1788 - 1797
Hlavní autoři: Liu, Zhi-Song, Siu, Wan-Chi, Wang, Li-Wen, Li, Chu-Tak, Cani, Marie-Paule, Chan, Yui-Lam
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.06.2020
Témata:
ISSN:2160-7516
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Benefited from the deep learning, image Super-Resolution has been one of the most developing research fields in computer vision. Depending upon whether using a discriminator or not, a deep convolutional neural network can provide an image with high fidelity or better perceptual quality. Due to the lack of ground truth images in real life, people prefer a photo-realistic image with low fidelity to a blurry image with high fidelity. In this paper, we revisit the classic example based image super-resolution approaches and come up with a novel generative model for perceptual image super-resolution. Given that real images contain various noise and artifacts, we propose a joint image denoising and super-resolution model via Variational AutoEncoder. We come up with a conditional variational autoencoder to encode the reference for dense feature vector which can then be transferred to the decoder for target image denoising. With the aid of the discriminator, an additional overhead of super-resolution subnetwork is attached to super-resolve the denoised image with photo-realistic visual quality. We participated the NTIRE2020 Real Image Super-Resolution Challenge [24] . Experimental results show that by using the proposed approach, we can obtain enlarged images with clean and pleasant features compared to other supervised methods. We also compared our approach with state-of-the-art methods on various datasets to demonstrate the efficiency of our proposed unsupervised super-resolution model.
AbstractList Benefited from the deep learning, image Super-Resolution has been one of the most developing research fields in computer vision. Depending upon whether using a discriminator or not, a deep convolutional neural network can provide an image with high fidelity or better perceptual quality. Due to the lack of ground truth images in real life, people prefer a photo-realistic image with low fidelity to a blurry image with high fidelity. In this paper, we revisit the classic example based image super-resolution approaches and come up with a novel generative model for perceptual image super-resolution. Given that real images contain various noise and artifacts, we propose a joint image denoising and super-resolution model via Variational AutoEncoder. We come up with a conditional variational autoencoder to encode the reference for dense feature vector which can then be transferred to the decoder for target image denoising. With the aid of the discriminator, an additional overhead of super-resolution subnetwork is attached to super-resolve the denoised image with photo-realistic visual quality. We participated the NTIRE2020 Real Image Super-Resolution Challenge [24] . Experimental results show that by using the proposed approach, we can obtain enlarged images with clean and pleasant features compared to other supervised methods. We also compared our approach with state-of-the-art methods on various datasets to demonstrate the efficiency of our proposed unsupervised super-resolution model.
Author Wang, Li-Wen
Chan, Yui-Lam
Liu, Zhi-Song
Li, Chu-Tak
Cani, Marie-Paule
Siu, Wan-Chi
Author_xml – sequence: 1
  givenname: Zhi-Song
  surname: Liu
  fullname: Liu, Zhi-Song
  organization: The Hong Kong Polytechnic University
– sequence: 2
  givenname: Wan-Chi
  surname: Siu
  fullname: Siu, Wan-Chi
  organization: The Hong Kong Polytechnic University
– sequence: 3
  givenname: Li-Wen
  surname: Wang
  fullname: Wang, Li-Wen
  organization: The Hong Kong Polytechnic University
– sequence: 4
  givenname: Chu-Tak
  surname: Li
  fullname: Li, Chu-Tak
  organization: The Hong Kong Polytechnic University
– sequence: 5
  givenname: Marie-Paule
  surname: Cani
  fullname: Cani, Marie-Paule
  organization: École polytechnique,LIX
– sequence: 6
  givenname: Yui-Lam
  surname: Chan
  fullname: Chan, Yui-Lam
  organization: The Hong Kong Polytechnic University
BookMark eNotj91Kw0AUhFdRsK19AhHyAonn7GY3u5cl1FooVKKtl2WzOZGVNCn5Kfj2TdGrgZn5BmbK7uqmJsaeESJEMC_p_j37khAbHXHgEAFwbm7YFBOu0QgF8pZNOCoIE4nqgc277gcAELSURkzYdld3w4nas--oCDKyVbA-2m8KPq5umFHXVEPvmzo4exusqKbW9v5Mwd623l6DkVgMfbOsXVNQ-8juS1t1NP_XGdu9Lj_Tt3CzXa3TxSb0HEQfckLkpcgVOQ0cea5i4WyuExe7pIx5EuucyqIwoJBya5yE8YzUI5BDSU7M2NPfrieiw6n1R9v-HgyOPanFBXYEUpw
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/CVPRW50498.2020.00229
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 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 1728193605
9781728193601
EISSN 2160-7516
EndPage 1797
ExternalDocumentID 9150958
Genre orig-research
GroupedDBID 6IE
6IF
6IL
6IN
AAJGR
AAWTH
ABLEC
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
M43
OCL
RIE
RIL
ID FETCH-LOGICAL-i203t-2e112f3b6ec80212b643cab87c4c7f42748befdd9061eba9c5093658b6eb0fec3
IEDL.DBID RIE
ISICitedReferencesCount 28
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000788279001107&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 02:30:40 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-2e112f3b6ec80212b643cab87c4c7f42748befdd9061eba9c5093658b6eb0fec3
PageCount 10
ParticipantIDs ieee_primary_9150958
PublicationCentury 2000
PublicationDate 2020-June
PublicationDateYYYYMMDD 2020-06-01
PublicationDate_xml – month: 06
  year: 2020
  text: 2020-June
PublicationDecade 2020
PublicationTitle IEEE Computer Society Conference on Computer Vision and Pattern Recognition workshops
PublicationTitleAbbrev CVPRW
PublicationYear 2020
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0001085593
Score 1.9825934
Snippet Benefited from the deep learning, image Super-Resolution has been one of the most developing research fields in computer vision. Depending upon whether using a...
SourceID ieee
SourceType Publisher
StartPage 1788
SubjectTerms Decoding
Distortion
Mathematical model
Noise reduction
Spatial resolution
Training
Title Unsupervised Real Image Super-Resolution via Generative Variational AutoEncoder
URI https://ieeexplore.ieee.org/document/9150958
WOSCitedRecordID wos000788279001107&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/eLvHCXMwlV1Lb8IwDLYA7bAT22DaWznsuA7oMzlOCLRdGGKDcUNp6kgcVhBt-f2z2wp22GW3yFJUNU7sL85nG-ARcWCtJesXKhk4vkqso1CjozQ_8-qBxSgpm01Ek4lcLtW0AU-HXBhELMln-MzD8i0_2ZiCQ2U9RehFBbIJzSgKq1ytYzyFCVfKq5N0Bn3VGy6ms6-AEDAzuFxmcLkMJH81USl9yLj9v6-fQfeYjCemBzdzDg1ML6Bdo0dRn82sA-_zNCu2fPQzks8IAIq3b7IW4oOlDsfpq10m9mstqnLTbOvEgq7LdUhQvBT5ZpRynvuuC_Px6HP46tTtEpy12_dyx0XCTtaLQzSSC7fHBDaMjmVkfBNZn66fMkabJIpcOMZaGfodjwAITYj7Fo13Ca10k-IVCGk1uXaduMaXvtGuChJLylNcGyhQGF9Dh9dnta0qYqzqpbn5W3wLp6yAimB1B618V-A9nJh9vs52D6UafwCJ0KBY
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT8JAEJ4gmugJFYxv9-DRCvRBu0dDJBARCQJyI9vtbMLBQijl9zvTNuDBi7fNJJumO7sz385-MwPwiNg0xpD1a8nAs1wZGUuiQksqfuZVTYN-lDWb8AeDYDaTwxI87XJhEDEjn-EzD7O3_GipUw6V1SWhF-kFB3DInbOKbK19RIUpV9Ip0nSaDVlvT4ejL48wMHO4bOZw2Qwlf7VRybxIp_K_759CbZ-OJ4Y7R3MGJYzPoVLgR1GczqQKH5M4SVd8-BOSjwgCit432QvxyVKLI_X5PhPbhRJ5wWm2dmJKF-YiKChe0s3yNeZM93UNJp3XcbtrFQ0TrIXdcDaWjYSejBO2UAdcuj0kuKFVGPja1b5x6QIahGiiSJITx1BJTb_jEAShCWHDoHYuoBwvY7wEERhFzl1FtnYDVytbepEh9UmuDuRJDK-gyuszX-U1MebF0lz_LX6A4-74vT_v9wZvN3DCysjpVrdQ3qxTvIMjvd0skvV9ptIfEzOjoQ
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%3Ajournal&rft.genre=proceeding&rft.title=IEEE+Computer+Society+Conference+on+Computer+Vision+and+Pattern+Recognition+workshops&rft.atitle=Unsupervised+Real+Image+Super-Resolution+via+Generative+Variational+AutoEncoder&rft.au=Liu%2C+Zhi-Song&rft.au=Siu%2C+Wan-Chi&rft.au=Wang%2C+Li-Wen&rft.au=Li%2C+Chu-Tak&rft.date=2020-06-01&rft.pub=IEEE&rft.eissn=2160-7516&rft.spage=1788&rft.epage=1797&rft_id=info:doi/10.1109%2FCVPRW50498.2020.00229&rft.externalDocID=9150958