Guided Dual Networks for Single Image Super-Resolution

Bibliographic Details
Title:	Guided Dual Networks for Single Image Super-Resolution
Authors:	Wenhui Chen, Chuangchuang Liu, Yitong Yan, Longcun Jin, Xianfang Sun, Xinyi Peng
Source:	IEEE Access, Vol 8, Pp 93608-93620 (2020)
Publisher Information:	IEEE, 2020.
Publication Year:	2020
Collection:	LCC:Electrical engineering. Electronics. Nuclear engineering
Subject Terms:	Convolutional neural network, dual network, generative adversarial network, single image super-resolution, Electrical engineering. Electronics. Nuclear engineering, TK1-9971
More Details:	The PSNR-oriented super-resolution (SR) methods pursue high reconstruction accuracy, but tend to produce over-smoothed results and lose plenty of high-frequency details. The GAN-based SR methods aim to generate more photo-realistic images, but the hallucinatory details are often accompanied with unsatisfying artifacts and noise. To address these problems, we propose a guided dual super-resolution network (GDSR), which exploits the advantages of both the PSNR-oriented and the GAN-based methods to achieve a good trade-off between reconstruction accuracy and perceptual quality. Specifically, our network contains two branches, where one is trained to extract global information and the other to focus on detail information. In this way, our network simultaneously generates SR images with high accuracy and satisfactory visual quality. To obtain more high-frequency features, we use the global features extracted from the low-frequency branch to guide the training of the high-frequency branch. Besides, our method utilizes a mask network to adaptively recover the final super-resolved image. Extensive experiments on several standard benchmarks show that our proposed method achieves better performance compared with state-of-the-art methods. The source code and the results of our GDSR are available at https://github.com/wenchen4321/GDSR.
Document Type:	article
File Description:	electronic resource
Language:	English
ISSN:	2169-3536
Relation:	https://ieeexplore.ieee.org/document/9097227/; https://doaj.org/toc/2169-3536
DOI:	10.1109/ACCESS.2020.2995175
Access URL:	https://doaj.org/article/c263d7417a5148b09af1cb847ab2a590
Accession Number:	edsdoj.263d7417a5148b09af1cb847ab2a590
Database:	Directory of Open Access Journals

FullText	Text: Availability: 0 CustomLinks: – Url: https://login.libproxy.scu.edu/login?url=http://ieeexplore.ieee.org/search/searchresult.jsp?action=search&newsearch=true&queryText=%22DOI%22:10.1109/ACCESS.2020.2995175 Name: EDS - IEEE (s8985755) Category: fullText Text: Check IEEE Xplore for full text MouseOverText: Check IEEE Xplore for full text. A new window will open. – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsdoj&genre=article&issn=21693536&ISBN=&volume=8&issue=&date=20200101&spage=93608&pages=93608-93620&title=IEEE Access&atitle=Guided%20Dual%20Networks%20for%20Single%20Image%20Super-Resolution&aulast=Wenhui%20Chen&id=DOI:10.1109/ACCESS.2020.2995175 Name: Full Text Finder (for New FTF UI) (s8985755) Category: fullText Text: Find It @ SCU Libraries MouseOverText: Find It @ SCU Libraries – Url: https://doaj.org/article/c263d7417a5148b09af1cb847ab2a590 Name: EDS - DOAJ (s8985755) Category: fullText Text: View record from DOAJ MouseOverText: View record from DOAJ
Header	DbId: edsdoj DbLabel: Directory of Open Access Journals An: edsdoj.263d7417a5148b09af1cb847ab2a590 RelevancyScore: 930 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 930.017944335938
IllustrationInfo
Items	– Name: Title Label: Title Group: Ti Data: Guided Dual Networks for Single Image Super-Resolution – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Wenhui+Chen%22">Wenhui Chen</searchLink><br /><searchLink fieldCode="AR" term="%22Chuangchuang+Liu%22">Chuangchuang Liu</searchLink><br /><searchLink fieldCode="AR" term="%22Yitong+Yan%22">Yitong Yan</searchLink><br /><searchLink fieldCode="AR" term="%22Longcun+Jin%22">Longcun Jin</searchLink><br /><searchLink fieldCode="AR" term="%22Xianfang+Sun%22">Xianfang Sun</searchLink><br /><searchLink fieldCode="AR" term="%22Xinyi+Peng%22">Xinyi Peng</searchLink> – Name: TitleSource Label: Source Group: Src Data: IEEE Access, Vol 8, Pp 93608-93620 (2020) – Name: Publisher Label: Publisher Information Group: PubInfo Data: IEEE, 2020. – Name: DatePubCY Label: Publication Year Group: Date Data: 2020 – Name: Subset Label: Collection Group: HoldingsInfo Data: LCC:Electrical engineering. Electronics. Nuclear engineering – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Convolutional+neural+network%22">Convolutional neural network</searchLink><br /><searchLink fieldCode="DE" term="%22dual+network%22">dual network</searchLink><br /><searchLink fieldCode="DE" term="%22generative+adversarial+network%22">generative adversarial network</searchLink><br /><searchLink fieldCode="DE" term="%22single+image+super-resolution%22">single image super-resolution</searchLink><br /><searchLink fieldCode="DE" term="%22Electrical+engineering%2E+Electronics%2E+Nuclear+engineering%22">Electrical engineering. Electronics. Nuclear engineering</searchLink><br /><searchLink fieldCode="DE" term="%22TK1-9971%22">TK1-9971</searchLink> – Name: Abstract Label: Description Group: Ab Data: The PSNR-oriented super-resolution (SR) methods pursue high reconstruction accuracy, but tend to produce over-smoothed results and lose plenty of high-frequency details. The GAN-based SR methods aim to generate more photo-realistic images, but the hallucinatory details are often accompanied with unsatisfying artifacts and noise. To address these problems, we propose a guided dual super-resolution network (GDSR), which exploits the advantages of both the PSNR-oriented and the GAN-based methods to achieve a good trade-off between reconstruction accuracy and perceptual quality. Specifically, our network contains two branches, where one is trained to extract global information and the other to focus on detail information. In this way, our network simultaneously generates SR images with high accuracy and satisfactory visual quality. To obtain more high-frequency features, we use the global features extracted from the low-frequency branch to guide the training of the high-frequency branch. Besides, our method utilizes a mask network to adaptively recover the final super-resolved image. Extensive experiments on several standard benchmarks show that our proposed method achieves better performance compared with state-of-the-art methods. The source code and the results of our GDSR are available at https://github.com/wenchen4321/GDSR. – Name: TypeDocument Label: Document Type Group: TypDoc Data: article – Name: Format Label: File Description Group: SrcInfo Data: electronic resource – Name: Language Label: Language Group: Lang Data: English – Name: ISSN Label: ISSN Group: ISSN Data: 2169-3536 – Name: NoteTitleSource Label: Relation Group: SrcInfo Data: https://ieeexplore.ieee.org/document/9097227/; https://doaj.org/toc/2169-3536 – Name: DOI Label: DOI Group: ID Data: 10.1109/ACCESS.2020.2995175 – Name: URL Label: Access URL Group: URL Data: <link linkTarget="URL" linkTerm="https://doaj.org/article/c263d7417a5148b09af1cb847ab2a590" linkWindow="_blank">https://doaj.org/article/c263d7417a5148b09af1cb847ab2a590</link> – Name: AN Label: Accession Number Group: ID Data: edsdoj.263d7417a5148b09af1cb847ab2a590
PLink	https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsdoj&AN=edsdoj.263d7417a5148b09af1cb847ab2a590
RecordInfo	BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1109/ACCESS.2020.2995175 Languages: – Text: English PhysicalDescription: Pagination: PageCount: 13 StartPage: 93608 Subjects: – SubjectFull: Convolutional neural network Type: general – SubjectFull: dual network Type: general – SubjectFull: generative adversarial network Type: general – SubjectFull: single image super-resolution Type: general – SubjectFull: Electrical engineering. Electronics. Nuclear engineering Type: general – SubjectFull: TK1-9971 Type: general Titles: – TitleFull: Guided Dual Networks for Single Image Super-Resolution Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Wenhui Chen – PersonEntity: Name: NameFull: Chuangchuang Liu – PersonEntity: Name: NameFull: Yitong Yan – PersonEntity: Name: NameFull: Longcun Jin – PersonEntity: Name: NameFull: Xianfang Sun – PersonEntity: Name: NameFull: Xinyi Peng IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2020 Identifiers: – Type: issn-print Value: 21693536 Numbering: – Type: volume Value: 8 Titles: – TitleFull: IEEE Access Type: main
ResultId	1