NTIRE 2017 Challenge on Single Image Super-Resolution: Methods and Results
Radu Timofte(ETH Zurich), Eirikur Agustsson(ETH Zurich), Luc Van Gool, Shuicheng Yan(University of California, Merced), Lei Zhang(Hong Kong Polytechnic University), Bee Lim(Seoul National University), Sanghyun Son(Seoul National University), Heewon Kim(Seoul National University), Seungjun Nah(Seoul National University), Kyoung Mu Lee(Seoul National University), Xintao Wang(University of Illinois Urbana-Champaign), Yapeng Tian(Shenzhen Institutes of Advanced Technology), Ke Yu(Chinese University of Hong Kong), Yulun Zhang(University of California, Merced), Shixiang Wu(Shenzhen Institutes of Advanced Technology), Chao Dong(Group Sense (China)), Liang Lin(Group Sense (China)), Yu Qiao(Shenzhen Institutes of Advanced Technology), Chen Change Loy(Chinese University of Hong Kong), Woong Bae(Korea Advanced Institute of Science and Technology), Jaejun Yoo(Korea Advanced Institute of Science and Technology), Yoseob Han(Korea Advanced Institute of Science and Technology), Jong Chul Ye(Korea Advanced Institute of Science and Technology), Jae-Seok Choi(Korea Advanced Institute of Science and Technology), Munchurl Kim(Korea Advanced Institute of Science and Technology), Yuchen Fan(University of Illinois Urbana-Champaign), Jiahui Yu(University of Illinois Urbana-Champaign), Wei Han(University of Illinois Urbana-Champaign), Ding Liu(University of Illinois Urbana-Champaign), Haichao Yu(University of Illinois Urbana-Champaign), Zhangyang Wang(University of Electronic Science and Technology of China), Honghui Shi(University of Illinois Urbana-Champaign), Xinchao Wang(University of Illinois Urbana-Champaign), Thomas S. Huang(University of Illinois Urbana-Champaign), Yunjin Chen(Université La Sagesse), Kai Zhang(Harbin Institute of Technology), Wangmeng Zuo(Harbin Institute of Technology), Zhimin Tang(Xiamen University), Linkai Luo(Xiamen University), Shaohui Li(Xiamen University), Min Fu(Xiamen University), Lei Cao(Xiamen University), Wen Heng(Peking University), Giang Bui(University of Missouri), Truc Le(University of Missouri), Ye Duan(University of Missouri), Dacheng Tao(The University of Sydney), Ruxin Wang, Lin Xu, Jianxin Pang, Jinchang Xu(Beijing University of Posts and Telecommunications), Yu Zhao(Beijing University of Posts and Telecommunications), Xiangyu Xu(University of California, Merced), Jinshan Pan(University of California, Merced), Deqing Sun(University of California, Merced), Yu‐Jin Zhang(University of California, Merced), Xibin Song(Shandong University), Yuchao Dai(Australian National University), Xueying Qin(Shandong University), Xuan-Phung Huynh(Sejong University), Tiantong Guo(Pennsylvania State University), Hojjat Seyed Mousavi(Pennsylvania State University), Tiep H. Vu(Pennsylvania State University), Vishal Monga(Pennsylvania State University), Cristóvão Cruz(Tampere University), Karen Egiazarian(Tampere University), Vladimir Katkovnik(Tampere University), Rakesh Mehta(Tampere University), Arnav Jain(Indian Institute of Technology Kharagpur), Abhinav Agarwalla(Indian Institute of Technology Kharagpur), Ch V. Sai Praveen(Indian Institute of Technology Kharagpur), Ruofan Zhou(École Polytechnique Fédérale de Lausanne), Hongdiao Wen(University of Electronic Science and Technology of China), Che Zhu(University of Electronic Science and Technology of China), Zhiqiang Xia(University of Electronic Science and Technology of China), Zhengtao Wang(University of Electronic Science and Technology of China), Qi Guo(University of Electronic Science and Technology of China)
Cited by 1,489Open Access
Abstract
This paper reviews the first challenge on single image super-resolution (restoration of rich details in an low resolution image) with focus on proposed solutions and results. A new DIVerse 2K resolution image dataset (DIV2K) was employed. The challenge had 6 competitions divided into 2 tracks with 3 magnification factors each. Track 1 employed the standard bicubic downscaling setup, while Track 2 had unknown downscaling operators (blur kernel and decimation) but learnable through low and high res train images. Each competition had ∽100 registered participants and 20 teams competed in the final testing phase. They gauge the state-of-the-art in single image super-resolution.
Related Papers
No related papers found
Powered by citation graph analysis