TY - GEN
T1 - AIM 2019 challenge on video extreme super-resolution
T2 - 17th IEEE/CVF International Conference on Computer Vision Workshop, ICCVW 2019
AU - Fuoli, Dario
AU - Gu, Shuhang
AU - Timofte, Radu
AU - Tao, Xin
AU - Li, Wenbo
AU - Guo, Taian
AU - Deng, Zijun
AU - Lu, Liying
AU - Dai, Tao
AU - Shen, Xiaoyong
AU - Xia, Shutao
AU - Dai, Yurong
AU - Jia, Jiaya
AU - Yi, Peng
AU - Wang, Zhongyuan
AU - Jiang, Kui
AU - Jiang, Junjun
AU - Ma, Jiayi
AU - Zhong, Zhiwei
AU - Wang, Chenyang
AU - Liu, Xianming
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/10
Y1 - 2019/10
N2 - This paper reviews the extreme video super-resolution challenge from the AIM 2019 workshop, with emphasis on submitted solutions and results. Video extreme super-resolution x16 is a highly challenging problem, because 256 pixels need to be estimated for each single pixel in the low-resolution (LR) input. Contrary to single image super-resolution (SISR), video provides temporal information, which can be additionally leveraged to restore the heavily downscaled videos and is imperative for any video super-resolution (VSR) method. The challenge is composed of two tracks, to find the best performing method for fully supervised VSR (track 1) and to find the solution which generates the perceptually best looking outputs (track 2). A new video dataset, called Vid3oC, is introduced together with the challenge.
AB - This paper reviews the extreme video super-resolution challenge from the AIM 2019 workshop, with emphasis on submitted solutions and results. Video extreme super-resolution x16 is a highly challenging problem, because 256 pixels need to be estimated for each single pixel in the low-resolution (LR) input. Contrary to single image super-resolution (SISR), video provides temporal information, which can be additionally leveraged to restore the heavily downscaled videos and is imperative for any video super-resolution (VSR) method. The challenge is composed of two tracks, to find the best performing method for fully supervised VSR (track 1) and to find the solution which generates the perceptually best looking outputs (track 2). A new video dataset, called Vid3oC, is introduced together with the challenge.
KW - AIM 2019
KW - Challenge
KW - Extreme video super resolution
KW - Super resoltuion
KW - Video super resolution
UR - https://www.scopus.com/pages/publications/85082460952
U2 - 10.1109/ICCVW.2019.00430
DO - 10.1109/ICCVW.2019.00430
M3 - 会议稿件
AN - SCOPUS:85082460952
T3 - Proceedings - 2019 International Conference on Computer Vision Workshop, ICCVW 2019
SP - 3467
EP - 3475
BT - Proceedings - 2019 International Conference on Computer Vision Workshop, ICCVW 2019
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 27 October 2019 through 28 October 2019
ER -