NTIRE 2020 challenge on real-world image super-resolution: Methods and results

Andreas Lugmayr; Martin Danelljan; Radu Timofte; Namhyuk Ahn; Dongwoon Bai; Jie Cai; Yun Cao; Junyang Chen; Kaihua Cheng; Se Young Chun; Wei Deng; Mostafa El-Khamy; Chiu Man Ho; Xiaozhong Ji; Amin Kheradmand; Gwantae Kim; Hanseok Ko; Kanghyu Lee; Jungwon Lee; Hao Li; Ziluan Liu; Zhi Song Liu; Shuai Liu; Yunhua Lu; Zibo Meng; Pablo Navarrete Michelini; Christian Micheloni; Kalpesh Prajapati; Haoyu Ren; Yong Hyeok Seo; Wan Chi Siu; Kyung Ah Sohn; Ying Tai; Rao Muhammad Umer; Shuangquan Wang; Huibing Wang; Timothy Haoning Wu; Haoning Wu; Biao Yang; Fuzhi Yang; Jaejun Yoo; Tongtong Zhao; Yuanbo Zhou; Haijie Zhuo; Ziyao Zong; Xueyi Zou; Li Wen Wang; Marie Paule Cani; Huan Yang; Jianlong Fu; Yukai Shi; Jaihyun Park; Junyeop Lee; Jeongki Min; Bokyeung Lee; Fengshuo Hu; Yanhong Wang; Jinjia Peng; Chengjie Wang; Jilin Li; Feiyue Huang; Xing Liu; Shuaijun Chen; Lei Zhao; Zhan Wang; Yuxuan Lin; Xu Jia; Qinquan Gao; Heena Patel; Vishal Chudasama; Kishor Upla; Raghavendra Ramachandra; Kiran Raja; Christoph Busch

doi:10.1109/CVPRW50498.2020.00255

NTIRE 2020 challenge on real-world image super-resolution: Methods and results

Andreas Lugmayr, Martin Danelljan, Radu Timofte, Namhyuk Ahn, Dongwoon Bai, Jie Cai, Yun Cao, Junyang Chen, Kaihua Cheng, Se Young Chun, Wei Deng, Mostafa El-Khamy, Chiu Man Ho, Xiaozhong Ji, Amin Kheradmand, Gwantae Kim, Hanseok Ko, Kanghyu Lee, Jungwon Lee, Hao LiZiluan Liu, Zhi Song Liu, Shuai Liu, Yunhua Lu, Zibo Meng, Pablo Navarrete Michelini, Christian Micheloni, Kalpesh Prajapati, Haoyu Ren, Yong Hyeok Seo, Wan Chi Siu, Kyung Ah Sohn, Ying Tai, Rao Muhammad Umer, Shuangquan Wang, Huibing Wang, Timothy Haoning Wu, Haoning Wu, Biao Yang, Fuzhi Yang, Jaejun Yoo, Tongtong Zhao, Yuanbo Zhou, Haijie Zhuo, Ziyao Zong, Xueyi Zou, Li Wen Wang, Marie Paule Cani, Huan Yang, Jianlong Fu, Yukai Shi, Jaihyun Park, Junyeop Lee, Jeongki Min, Bokyeung Lee, Fengshuo Hu, Yanhong Wang, Jinjia Peng, Chengjie Wang, Jilin Li, Feiyue Huang, Xing Liu, Shuaijun Chen, Lei Zhao, Zhan Wang, Yuxuan Lin, Xu Jia, Qinquan Gao, Heena Patel, Vishal Chudasama, Kishor Upla, Raghavendra Ramachandra, Kiran Raja, Christoph Busch

School of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

141 Citations (Scopus)

Abstract

This paper reviews the NTIRE 2020 challenge on real world super-resolution. It focuses on the participating methods and final results. The challenge addresses the real world setting, where paired true high and low-resolution images are unavailable. For training, only one set of source input images is therefore provided along with a set of unpaired high-quality target images. In Track 1: Image Processing artifacts, the aim is to super-resolve images with synthetically generated image processing artifacts. This allows for quantitative benchmarking of the approaches w.r.t. a ground-truth image. In Track 2: Smartphone Images, real low-quality smart phone images have to be super-resolved. In both tracks, the ultimate goal is to achieve the best perceptual quality, evaluated using a human study. This is the second challenge on the subject, following AIM 2019, targeting to advance the state-of-the-art in super-resolution. To measure the performance we use the benchmark protocol from AIM 2019. In total 22 teams competed in the final testing phase, demonstrating new and innovative solutions to the problem.

Original language	English
Title of host publication	Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020
Publisher	IEEE Computer Society
Pages	2058-2076
Number of pages	19
ISBN (Electronic)	9781728193601
DOIs	https://doi.org/10.1109/CVPRW50498.2020.00255
Publication status	Published - 2020 Jun
Event	2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020 - Virtual, Online, United States Duration: 2020 Jun 14 → 2020 Jun 19

Publication series

Name	IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Volume	2020-June
ISSN (Print)	2160-7508
ISSN (Electronic)	2160-7516

Conference

Conference	2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020
Country/Territory	United States
City	Virtual, Online
Period	20/6/14 → 20/6/19

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Electrical and Electronic Engineering

Access to Document

10.1109/CVPRW50498.2020.00255

Cite this

Lugmayr, A., Danelljan, M., Timofte, R., Ahn, N., Bai, D., Cai, J., Cao, Y., Chen, J., Cheng, K., Chun, S. Y., Deng, W., El-Khamy, M., Ho, C. M., Ji, X., Kheradmand, A., Kim, G., Ko, H., Lee, K., Lee, J., ... Busch, C. (2020). NTIRE 2020 challenge on real-world image super-resolution: Methods and results. In Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020 (pp. 2058-2076). Article 9150677 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Vol. 2020-June). IEEE Computer Society. https://doi.org/10.1109/CVPRW50498.2020.00255

NTIRE 2020 challenge on real-world image super-resolution: Methods and results. / Lugmayr, Andreas; Danelljan, Martin; Timofte, Radu et al.
Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020. IEEE Computer Society, 2020. p. 2058-2076 9150677 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Vol. 2020-June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Lugmayr, A, Danelljan, M, Timofte, R, Ahn, N, Bai, D, Cai, J, Cao, Y, Chen, J, Cheng, K, Chun, SY, Deng, W, El-Khamy, M, Ho, CM, Ji, X, Kheradmand, A, Kim, G, Ko, H, Lee, K, Lee, J, Li, H, Liu, Z, Liu, ZS, Liu, S, Lu, Y, Meng, Z, Michelini, PN, Micheloni, C, Prajapati, K, Ren, H, Seo, YH, Siu, WC, Sohn, KA, Tai, Y, Umer, RM, Wang, S, Wang, H, Wu, TH, Wu, H, Yang, B, Yang, F, Yoo, J, Zhao, T, Zhou, Y, Zhuo, H, Zong, Z, Zou, X, Wang, LW, Cani, MP, Yang, H, Fu, J, Shi, Y, Park, J, Lee, J, Min, J, Lee, B, Hu, F, Wang, Y, Peng, J, Wang, C, Li, J, Huang, F, Liu, X, Chen, S, Zhao, L, Wang, Z, Lin, Y, Jia, X, Gao, Q, Patel, H, Chudasama, V, Upla, K, Ramachandra, R, Raja, K & Busch, C 2020, NTIRE 2020 challenge on real-world image super-resolution: Methods and results. in Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020., 9150677, IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, vol. 2020-June, IEEE Computer Society, pp. 2058-2076, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020, Virtual, Online, United States, 20/6/14. https://doi.org/10.1109/CVPRW50498.2020.00255

Lugmayr A, Danelljan M, Timofte R, Ahn N, Bai D, Cai J et al. NTIRE 2020 challenge on real-world image super-resolution: Methods and results. In Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020. IEEE Computer Society. 2020. p. 2058-2076. 9150677. (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops). doi: 10.1109/CVPRW50498.2020.00255

Lugmayr, Andreas ; Danelljan, Martin ; Timofte, Radu et al. / NTIRE 2020 challenge on real-world image super-resolution : Methods and results. Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020. IEEE Computer Society, 2020. pp. 2058-2076 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).

@inproceedings{5ea372578d4f4505a7c5c12b8de379c7,

title = "NTIRE 2020 challenge on real-world image super-resolution: Methods and results",

abstract = "This paper reviews the NTIRE 2020 challenge on real world super-resolution. It focuses on the participating methods and final results. The challenge addresses the real world setting, where paired true high and low-resolution images are unavailable. For training, only one set of source input images is therefore provided along with a set of unpaired high-quality target images. In Track 1: Image Processing artifacts, the aim is to super-resolve images with synthetically generated image processing artifacts. This allows for quantitative benchmarking of the approaches w.r.t. a ground-truth image. In Track 2: Smartphone Images, real low-quality smart phone images have to be super-resolved. In both tracks, the ultimate goal is to achieve the best perceptual quality, evaluated using a human study. This is the second challenge on the subject, following AIM 2019, targeting to advance the state-of-the-art in super-resolution. To measure the performance we use the benchmark protocol from AIM 2019. In total 22 teams competed in the final testing phase, demonstrating new and innovative solutions to the problem.",

author = "Andreas Lugmayr and Martin Danelljan and Radu Timofte and Namhyuk Ahn and Dongwoon Bai and Jie Cai and Yun Cao and Junyang Chen and Kaihua Cheng and Chun, {Se Young} and Wei Deng and Mostafa El-Khamy and Ho, {Chiu Man} and Xiaozhong Ji and Amin Kheradmand and Gwantae Kim and Hanseok Ko and Kanghyu Lee and Jungwon Lee and Hao Li and Ziluan Liu and Liu, {Zhi Song} and Shuai Liu and Yunhua Lu and Zibo Meng and Michelini, {Pablo Navarrete} and Christian Micheloni and Kalpesh Prajapati and Haoyu Ren and Seo, {Yong Hyeok} and Siu, {Wan Chi} and Sohn, {Kyung Ah} and Ying Tai and Umer, {Rao Muhammad} and Shuangquan Wang and Huibing Wang and Wu, {Timothy Haoning} and Haoning Wu and Biao Yang and Fuzhi Yang and Jaejun Yoo and Tongtong Zhao and Yuanbo Zhou and Haijie Zhuo and Ziyao Zong and Xueyi Zou and Wang, {Li Wen} and Cani, {Marie Paule} and Huan Yang and Jianlong Fu and Yukai Shi and Jaihyun Park and Junyeop Lee and Jeongki Min and Bokyeung Lee and Fengshuo Hu and Yanhong Wang and Jinjia Peng and Chengjie Wang and Jilin Li and Feiyue Huang and Xing Liu and Shuaijun Chen and Lei Zhao and Zhan Wang and Yuxuan Lin and Xu Jia and Qinquan Gao and Heena Patel and Vishal Chudasama and Kishor Upla and Raghavendra Ramachandra and Kiran Raja and Christoph Busch",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020 ; Conference date: 14-06-2020 Through 19-06-2020",

year = "2020",

month = jun,

doi = "10.1109/CVPRW50498.2020.00255",

language = "English",

series = "IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",

publisher = "IEEE Computer Society",

pages = "2058--2076",

booktitle = "Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020",

}

TY - GEN

T1 - NTIRE 2020 challenge on real-world image super-resolution

T2 - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020

AU - Lugmayr, Andreas

AU - Danelljan, Martin

AU - Timofte, Radu

AU - Ahn, Namhyuk

AU - Bai, Dongwoon

AU - Cai, Jie

AU - Cao, Yun

AU - Chen, Junyang

AU - Cheng, Kaihua

AU - Chun, Se Young

AU - Deng, Wei

AU - El-Khamy, Mostafa

AU - Ho, Chiu Man

AU - Ji, Xiaozhong

AU - Kheradmand, Amin

AU - Kim, Gwantae

AU - Ko, Hanseok

AU - Lee, Kanghyu

AU - Lee, Jungwon

AU - Li, Hao

AU - Liu, Ziluan

AU - Liu, Zhi Song

AU - Liu, Shuai

AU - Lu, Yunhua

AU - Meng, Zibo

AU - Michelini, Pablo Navarrete

AU - Micheloni, Christian

AU - Prajapati, Kalpesh

AU - Ren, Haoyu

AU - Seo, Yong Hyeok

AU - Siu, Wan Chi

AU - Sohn, Kyung Ah

AU - Tai, Ying

AU - Umer, Rao Muhammad

AU - Wang, Shuangquan

AU - Wang, Huibing

AU - Wu, Timothy Haoning

AU - Wu, Haoning

AU - Yang, Biao

AU - Yang, Fuzhi

AU - Yoo, Jaejun

AU - Zhao, Tongtong

AU - Zhou, Yuanbo

AU - Zhuo, Haijie

AU - Zong, Ziyao

AU - Zou, Xueyi

AU - Wang, Li Wen

AU - Cani, Marie Paule

AU - Yang, Huan

AU - Fu, Jianlong

AU - Shi, Yukai

AU - Park, Jaihyun

AU - Lee, Junyeop

AU - Min, Jeongki

AU - Lee, Bokyeung

AU - Hu, Fengshuo

AU - Wang, Yanhong

AU - Peng, Jinjia

AU - Wang, Chengjie

AU - Li, Jilin

AU - Huang, Feiyue

AU - Liu, Xing

AU - Chen, Shuaijun

AU - Zhao, Lei

AU - Wang, Zhan

AU - Lin, Yuxuan

AU - Jia, Xu

AU - Gao, Qinquan

AU - Patel, Heena

AU - Chudasama, Vishal

AU - Upla, Kishor

AU - Ramachandra, Raghavendra

AU - Raja, Kiran

AU - Busch, Christoph

PY - 2020/6

Y1 - 2020/6

N2 - This paper reviews the NTIRE 2020 challenge on real world super-resolution. It focuses on the participating methods and final results. The challenge addresses the real world setting, where paired true high and low-resolution images are unavailable. For training, only one set of source input images is therefore provided along with a set of unpaired high-quality target images. In Track 1: Image Processing artifacts, the aim is to super-resolve images with synthetically generated image processing artifacts. This allows for quantitative benchmarking of the approaches w.r.t. a ground-truth image. In Track 2: Smartphone Images, real low-quality smart phone images have to be super-resolved. In both tracks, the ultimate goal is to achieve the best perceptual quality, evaluated using a human study. This is the second challenge on the subject, following AIM 2019, targeting to advance the state-of-the-art in super-resolution. To measure the performance we use the benchmark protocol from AIM 2019. In total 22 teams competed in the final testing phase, demonstrating new and innovative solutions to the problem.

AB - This paper reviews the NTIRE 2020 challenge on real world super-resolution. It focuses on the participating methods and final results. The challenge addresses the real world setting, where paired true high and low-resolution images are unavailable. For training, only one set of source input images is therefore provided along with a set of unpaired high-quality target images. In Track 1: Image Processing artifacts, the aim is to super-resolve images with synthetically generated image processing artifacts. This allows for quantitative benchmarking of the approaches w.r.t. a ground-truth image. In Track 2: Smartphone Images, real low-quality smart phone images have to be super-resolved. In both tracks, the ultimate goal is to achieve the best perceptual quality, evaluated using a human study. This is the second challenge on the subject, following AIM 2019, targeting to advance the state-of-the-art in super-resolution. To measure the performance we use the benchmark protocol from AIM 2019. In total 22 teams competed in the final testing phase, demonstrating new and innovative solutions to the problem.

UR - http://www.scopus.com/inward/record.url?scp=85090110747&partnerID=8YFLogxK

U2 - 10.1109/CVPRW50498.2020.00255

DO - 10.1109/CVPRW50498.2020.00255

M3 - Conference contribution

AN - SCOPUS:85090110747

T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

SP - 2058

EP - 2076

BT - Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020

PB - IEEE Computer Society

Y2 - 14 June 2020 through 19 June 2020

ER -

NTIRE 2020 challenge on real-world image super-resolution: Methods and results

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this