AIM 2020 高效超分辨率挑战：方法与结果
图像与视频处理
2020-09-16 v1 计算机视觉与模式识别
摘要
本文回顾了 AIM 2020 高效单图像超分辨率挑战，重点关注所提出的解决方案与结果。挑战任务是基于一组低分辨率与对应高分辨率图像的先验样例，以放大系数 x4 对输入图像进行超分辨率重建。目标是设计一种网络，在降低运行时、参数数量、FLOPs、激活值与内存消耗等一个或多个方面的同时，至少保持 MSRResNet 的 PSNR。该赛道有 150 名注册参与者，25 支队伍提交了最终结果。它们衡量了高效单图像超分辨率的最先进水平。
引用
@article{arxiv.2009.06943,
title = {AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results},
author = {Kai Zhang and Martin Danelljan and Yawei Li and Radu Timofte and Jie Liu and Jie Tang and Gangshan Wu and Yu Zhu and Xiangyu He and Wenjie Xu and Chenghua Li and Cong Leng and Jian Cheng and Guangyang Wu and Wenyi Wang and Xiaohong Liu and Hengyuan Zhao and Xiangtao Kong and Jingwen He and Yu Qiao and Chao Dong and Xiaotong Luo and Liang Chen and Jiangtao Zhang and Maitreya Suin and Kuldeep Purohit and A. N. Rajagopalan and Xiaochuan Li and Zhiqiang Lang and Jiangtao Nie and Wei Wei and Lei Zhang and Abdul Muqeet and Jiwon Hwang and Subin Yang and JungHeum Kang and Sung-Ho Bae and Yongwoo Kim and Liang Chen and Jiangtao Zhang and Xiaotong Luo and Yanyun Qu and Geun-Woo Jeon and Jun-Ho Choi and Jun-Hyuk Kim and Jong-Seok Lee and Steven Marty and Eric Marty and Dongliang Xiong and Siang Chen and Lin Zha and Jiande Jiang and Xinbo Gao and Wen Lu and Haicheng Wang and Vineeth Bhaskara and Alex Levinshtein and Stavros Tsogkas and Allan Jepson and Xiangzhen Kong and Tongtong Zhao and Shanshan Zhao and Hrishikesh P S and Densen Puthussery and Jiji C and Nan Nan and Shuai Liu and Jie Cai and Zibo Meng and Jiaming Ding and Chiu Man Ho and Xuehui Wang and Qiong Yan and Yuzhi Zhao and Long Chen and Jiangtao Zhang and Xiaotong Luo and Liang Chen and Yanyun Qu and Long Sun and Wenhao Wang and Zhenbing Liu and Rushi Lan and Rao Muhammad Umer and Christian Micheloni},
journal= {arXiv preprint arXiv:2009.06943},
year = {2020}
}