AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results

被引：56

作者：

Zhang, Kai ^{[1
]}

Danelljan, Martin ^{[1
]}

Li, Yawei ^{[1
]}

Timofte, Radu

Liu, Jie ^{[2
]}

Tang, Jie ^{[2
]}

Wu, Gangshan ^{[2
]}

Zhu, Yu ^{[3
]}

He, Xiangyu ^{[3
]}

Xu, Wenjie ^{[3
]}

Li, Chenghua ^{[3
]}

Leng, Cong ^{[3
]}

Cheng, Jian ^{[3
]}

Wu, Guangyang ^{[4
]}

Wang, Wenyi ^{[4
]}

Liu, Xiaohong ^{[5
]}

Zhao, Hengyuan ^{[6
]}

Kong, Xiangtao ^{[6
]}

He, Jingwen ^{[6
]}

Qiao, Yu ^{[6
]}

Dong, Chao ^{[6
]}

Luo, Xiaotong ^{[7
]}

Chen, Liang ^{[7
]}

Zhang, Jiangtao ^{[7
]}

Suin, Maitreya ^{[8
]}

Purohit, Kuldeep ^{[8
]}

Rajagopalan, A. N. ^{[8
]}

Li, Xiaochuan ^{[9
]}

Lang, Zhiqiang ^{[10
]}

Nie, Jiangtao ^{[10
]}

Wei, Wei ^{[10
]}

Zhang, Lei ^{[10
]}

Muqeet, Abdul ^{[11
]}

Hwang, Jiwon ^{[11
]}

Yang, Subin ^{[11
]}

Kang, JungHeum ^{[11
]}

Bae, Sung-Ho ^{[11
]}

Kim, Yongwoo ^{[12
]}

Qu, Yanyun

Jeon, Geun-Woo ^{[13
]}

Choi, Jun-Ho ^{[13
]}

Kim, Jun-Hyuk ^{[13
]}

Lee, Jong-Seok ^{[13
]}

Marty, Steven ^{[14
]}

Marty, Eric ^{[14
]}

Xiong, Dongliang ^{[15
]}

Chen, Siang ^{[15
]}

Zha, Lin ^{[16
]}

Jiang, Jiande ^{[16
]}

Gao, Xinbo ^{[17
]}

机构：

[1] Swiss Fed Inst Technol, Comp Vis Lab, Zurich, Switzerland

[2] Nanjing Univ, State Key Lab Novel Software Technol, Nanjing 210023, Peoples R China

[3] Chinese Acad Sci, AiRiA, Nanjing Artificial Intelligence Chip Res Inst Aut, Nanjing, Peoples R China

[4] Univ Elect Sci & Technol China, Nanjing, Peoples R China

[5] McMaster Univ, Hamilton, ON, Canada

[6] Chinese Acad Sci, Shenzhen Inst Adv Technol, Beijing, Peoples R China

[7] Xiamen Univ, Xiamen, Peoples R China

[8] Indian Inst Technol Madras, Chennai, Tamil Nadu, India

[9] Nanjing Univ Aeronaut & Astronaut, Nanjing, Peoples R China

[10] Northwestern Polytech Univ, Sch Comp Sci, Xian, Peoples R China

[11] Kyung Hee Univ, Seoul, South Korea

[12] Sang Myung Univ, Seoul, South Korea

[13] Yonsei Univ, Seoul, South Korea

[14] Swiss Fed Inst Technol, Zurich, Switzerland

[15] Zhejiang Univ, Hangzhou, Peoples R China

[16] Qingdao Hi Image Technol Co Ltd, Hisense Visual Technol Co, Shandong, Peoples R China

[17] Xidian Univ, Xian, Peoples R China

[18] Samsung Ctr, Toronto, ON, Canada

[19] Dalian Maritime Univ, Dalian, Peoples R China

[20] China Everbright Bank Co Ltd, Beijing, Peoples R China

[21] Coll Engn, Trivandrum, Kerala, India

[22] InnoPeak Technol Inc, Palo Alto, CA USA

[23] SenseTime Res, Hong Kong, Peoples R China

[24] Sun Yat Sen Univ, Guangzhou, Peoples R China

[25] City Univ Hong Kong, Kowloon, Hong Kong, Peoples R China

[26] Guilin Univ Elect Technol, Guilin 541004, Peoples R China

[27] Univ Udine, Udine, Italy

来源：

COMPUTER VISION - ECCV 2020 WORKSHOPS, PT III | 2020年 / 12537卷

关键词：

D O I：

10.1007/978-3-030-67070-2_1

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper reviews the AIM 2020 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The challenge task was to super-resolve an input image with a magnification factor x4 based on a set of prior examples of low and corresponding high resolution images. The goal is to devise a network that reduces one or several aspects such as runtime, parameter count, FLOPs, activations, and memory consumption while at least maintaining PSNR of MSRResNet. The track had 150 registered participants, and 25 teams submitted the final results. They gauge the state-of-the-art in efficient single image super-resolution.

引用

页码：5 / 40

页数：36

共 65 条

[1] NTIRE 2017 Challenge on Single Image Super-Resolution: Dataset and Study [J].

Agustsson, Eirikur ;

Timofte, Radu .

2017 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW), 2017, :1122-1131

[2]

[Anonymous], 2010, JMLR WORKSH C P

[3] Explorable Super Resolution [J].

Bahat, Yuval ;

Michaeli, Tomer .

2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, :2713-2722

[4]

Bell-Kligler S, 2019, ADV NEUR IN, V32

[5]

Cai H, 2020, Arxiv, DOI arXiv:1908.09791

[6]

Cai YH, 2020, Arxiv, DOI arXiv:2003.04030

[7] Adaptive Hybrid Composition Based Super-Resolution Network via Fine-Grained Channel Pruning [J].

Chen, Siang ;

Huang, Kai ;

Li, Bowen ;

Xiong, Dongliang ;

Jiang, Haitian ;

Claesen, Luc .

COMPUTER VISION - ECCV 2020 WORKSHOPS, PT III, 2020, 12537 :119-135

[8] ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks [J].

Ding, Xiaohan ;

Guo, Yuchen ;

Ding, Guiguang ;

Han, Jungong .

2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :1911-1920

[9] Centripetal SGD for Pruning Very Deep Convolutional Networks With Complicated Structure [J].

Ding, Xiaohan ;

Ding, Guiguang ;

Guo, Yuchen ;

Han, Jungong .

2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, :4938-4948

[10] Learning a Deep Convolutional Network for Image Super-Resolution [J].

Dong, Chao ;

Loy, Chen Change ;

He, Kaiming ;

Tang, Xiaoou .

COMPUTER VISION - ECCV 2014, PT IV, 2014, 8692 :184-199

← 1 2 3 4 5 6 7 →