AIM 2022 压缩图像与视频超分辨率挑战：数据集、方法与结果
图像与视频处理
2022-08-26 v2 计算机视觉与模式识别
摘要
本文回顾了 AIM 2022 上的压缩图像与视频超分辨率挑战。该挑战包含两条赛道。赛道 1 旨在压缩图像超分辨率，赛道 2 面向压缩视频超分辨率。在赛道 1 中，我们使用流行的 DIV2K 数据集作为训练、验证和测试集。在赛道 2 中，我们提出了 LDV 3.0 数据集，其包含 365 个视频，包括 LDV 2.0 数据集（335 个视频）和 30 个额外视频。在本挑战中，分别有 12 支和 2 支队伍向赛道 1 和赛道 2 提交了最终结果。所提出的方法与解决方案衡量了压缩图像与视频超分辨率的当前最优水平。所提出的 LDV 3.0 数据集可在 https://github.com/RenYang-home/LDV_dataset 获取。本挑战主页位于 https://github.com/RenYang-home/AIM22_CompressSR。
引用
@article{arxiv.2208.11184,
title = {AIM 2022 Challenge on Super-Resolution of Compressed Image and Video: Dataset, Methods and Results},
author = {Ren Yang and Radu Timofte and Xin Li and Qi Zhang and Lin Zhang and Fanglong Liu and Dongliang He and Fu li and He Zheng and Weihang Yuan and Pavel Ostyakov and Dmitry Vyal and Magauiya Zhussip and Xueyi Zou and Youliang Yan and Lei Li and Jingzhu Tang and Ming Chen and Shijie Zhao and Yu Zhu and Xiaoran Qin and Chenghua Li and Cong Leng and Jian Cheng and Claudio Rota and Marco Buzzelli and Simone Bianco and Raimondo Schettini and Dafeng Zhang and Feiyu Huang and Shizhuo Liu and Xiaobing Wang and Zhezhu Jin and Bingchen Li and Xin Li and Mingxi Li and Ding Liu and Wenbin Zou and Peijie Dong and Tian Ye and Yunchen Zhang and Ming Tan and Xin Niu and Mustafa Ayazoglu and Marcos Conde and Ui-Jin Choi and Zhuang Jia and Tianyu Xu and Yijian Zhang and Mao Ye and Dengyan Luo and Xiaofeng Pan and Liuhan Peng},
journal= {arXiv preprint arXiv:2208.11184},
year = {2022}
}
备注
Camera-ready version