There are no files associated with this item.
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.citation.conferencePlace | US | - |
dc.citation.conferencePlace | Salt Lake City | - |
dc.citation.endPage | 1003 | - |
dc.citation.startPage | 995 | - |
dc.citation.title | 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018 | - |
dc.contributor.author | Park, Dongwon | - |
dc.contributor.author | Kim, Kwanyoung | - |
dc.contributor.author | Chun, Se Young | - |
dc.date.accessioned | 2023-12-19T15:47:59Z | - |
dc.date.available | 2023-12-19T15:47:59Z | - |
dc.date.created | 2018-06-29 | - |
dc.date.issued | 2018-06-18 | - |
dc.description.abstract | Example based single image super resolution (SR) is a fundamental task in computer vision. It is challenging, but recently, there have been significant performance improve-ments using deep learning approaches. In this article, we propose efficient module based single image SR networks (EMBSR) and tackle multiple SR problems in NTIRE 2018 SR challenge by recycling trained networks. Our proposed EMBSR allowed us to reduce training time with effectively deeper networks, to use modular ensemble for improved performance, and to separate subproblems for better per-formance. We also proposed EDSR-PP, an improved ver-sion of previous ESDR by incorporating pyramid pooling so that global as well as local context information can be utilized. Lastly, we proposed a novel denoising / deblurring residual convolutional network (DnResNet) using residual block and batch normalization. Our proposed EMBSR with DnResNet and EDSR-PP demonstrated that multiple SR problems can be tackled efficiently and effectively by win-ning the 2nd place for Track 2 (×4 SR with mild adverse condition) and the 3rd place for Track 3 (×4 SR with diffi-cult adverse condition). Our proposed method with EDSR-PP also achieved the ninth place for Track 1 (×8 SR) with the fastest run time among top nine teams. |
- |
dc.identifier.bibliographicCitation | 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018, pp.995 - 1003 | - |
dc.identifier.doi | 10.1109/CVPRW.2018.00133 | - |
dc.identifier.issn | 2160-7508 | - |
dc.identifier.scopusid | 2-s2.0-85060871684 | - |
dc.identifier.uri | https://scholarworks.unist.ac.kr/handle/201301/35197 | - |
dc.identifier.url | https://ieeexplore.ieee.org/document/8575286 | - |
dc.language | 영어 | - |
dc.publisher | IEEE Computer Society | - |
dc.title | Efficient Module Based Single Image Super Resolution for Multiple Problems | - |
dc.type | Conference Paper | - |
dc.date.conferenceDate | 2018-06-18 | - |
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Tel : 052-217-1404 / Email : scholarworks@unist.ac.kr
Copyright (c) 2023 by UNIST LIBRARY. All rights reserved.
ScholarWorks@UNIST was established as an OAK Project for the National Library of Korea.