文献类型：会议论文

英文题名：NTIRE 2021 NonHomogeneous Dehazing Challenge Report

作者：Ancuti, Codruta O.[1];Ancuti, Cosmin;Vasluianu, Florin-Alexandru[1,2];Timofte, Radu[1,2];Fu, Minghan;Liu, Huan;Yu, Yankun[3];Chen, Jun[3];Wang, Keyan[4];Chang, Jerome[5];Wang, Xiyao[6];Liu, Jing[7];Xu, Yi[8];Zhang, Xinjian;Zhao, Minyi;Zhou, Shuigeng;Chen, Tianyi[9];Fu, Jiahui[9];Jiang, Wentao[9];Gao, Chen[9];Liu, Si[9];Wang, Yudong[10];Guo, Jichang[10];Li, Chongyi[11];Yan, Qixin[10];Zheng, Sida[10];Zamir, Syed Waqas[12];Arora, Aditya[12];Dudhane, Akshay[12];Khan, Salman[12];Hayat, Munawar[12];Khan, Fahad Shahbaz[12];Shao, Ling[12];Zhang, Haichuan[13];Guo, Tiantong[13];Monga, Vishal[13];Yang, Wenjin[14];Lin, Jin[14];Luo, Xiaotong[14];Huang, Guowen[14];Chen, Shuxin[14];Qu, Yanyun[14];Xu, Kele[15];Yang, Lehan[15];Sun, Pengliang[15];Niu, Xuetong[15];Zheng, Junjun[16];Ruan, Xiaotong[16];Wang, Yunfeng[16];Yang, Jiang[16];Luo, Zhipeng[17];Wang, Sai[17];Xu, Zhenyu[17];Cao, Xiaochun[18];Luo, Jun[18];Zheng, Zhuoran[18];Ren, Wenqi[18];Wang, Tao[18];Chen, Yiqun[19];Leng, Cong[20];Li, Chenghua[19];Cheng, Jian[20];Sung, Chang-Sung[21];Chen, Jun-Cheng[21];Jo, Eunsung[22];Sim, Jae-Young;Geethu, M. M.[23];Akhil, K. A.;Sreeni, K. G.[24];Jeena, R. S.;Zacharias, Joseph;Manu, Chippy M.[24];Huang, Zexi[25];Zhang, Baofeng[25];Zhang, Yiwen[25];Li, Jindong[25];Chen, Mianjie[25];Xiao, Quan[26];Su, Qingchao[26];Han, Lihua[26];Huang, Yanting[26];Prajapati, Kalpesh[27];Chudasama, Vishal[27];Patel, Heena[27];Sarvaiya, Anjali[27];Upla, Kishor[27];Raja, Kiran[27];Ramachandra, Raghavendra[27];Busch, Christoph[27];Jing, Hongyuan[28];Huang, Zilong[28];Fu, Yiran[28];Wu, Haoqiang[28];Zha, Quanxing[28];Zhu, Zhiwei[28];Lv, Hejun[28]

第一作者：Ancuti, Codruta O.

通讯作者：Ancuti, CO[1]

机构：[1]Univ Politehn Timisoara, ETcTI, Timisoara, Romania; UCL, ICTEAM, Louvain, Belgium;[2]Swiss Fed Inst Technol, Comp Vis Lab, Zurich, Switzerland;[3]McMaster Univ, Hamilton, ON, Canada;[4]Xidian Univ, Xian, Peoples R China;[5]Natl Taiwan Univ, Taipei, Taiwan;[6]Univ Chinese Acad Sci, Inst Automat, Ctr Res Intelligent Syst & Engn, Beijing, Peoples R China;[7]Bilibili Inc, Shanghai, Peoples R China;[8]Fudan Univ, Shanghai, Peoples R China;[9]Beihang Univ, Beijing, Peoples R China;[10]Tianjin Univ, Sch Elect & Informat Engn, Tianjin 300072, Peoples R China;[11]Nanyang Technol Univ NTU, Sch Comp Sci & Engn, Singapore, Singapore;[12]Incept Inst Artificial Intelligence IIAI, Abu Dhabi, U Arab Emirates;[13]Penn State Univ, Sch Elect Engn & Comp Sci, Informat Proc & Algorithms Lab iPAL, University Pk, PA 16802 USA;[14]Xiamen Univ, Dept Comp Sci, Xiamen, Peoples R China;[15]Key Lab Parallel & Distributed Proc, Changsha, Peoples R China;[16]Alibaba Inc, Shanghai, Peoples R China;[17]DeepBlue Technol Shanghai Co Ltd, Shanghai, Peoples R China;[18]Chinese Acad Sci, Inst Informat Engn, State Key Lab Informat Secur, Beijing, Peoples R China;[19]CASIA, Beijing, Peoples R China;[20]CASIA, AiRiA, Beijing, Peoples R China;[21]Natl Taiwan Univ, Data Sci Degree Program, Taipei, Taiwan;[22]Ulsan Natl Inst Sci & Technol, Ulsan, South Korea;[23]Coll Engn, Dept ECE, CV Lab, Trivandrum, Kerala, India;[24]AICTE Univ, Kollam, Kerala, India;[25]South China Univ Techonol, Sch Elect & Informat Engn, Guangzhou, Peoples R China;[26]Shenzhen Wave Kingdom Co Ltd, Shenzhen, Peoples R China;[27]Sardar Vallabhbhai Natl Inst Technol, Surat, India;[28]Beijing Union Univ, Coll Robot, Beijing, Peoples R China

第一机构：Univ Politehn Timisoara, ETcTI, Timisoara, Romania,UCL, ICTEAM, Louvain, Belgium

通讯机构：[1]corresponding author), Univ Politehn Timisoara, ETcTI, Timisoara, Romania.

会议论文集：IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

会议日期：JUN 19-25, 2021

会议地点：ELECTR NETWORK

语种：英文

摘要：This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. The nonhomogeneous haze has been introduced in the outdoor scenes by using a a professional setup that imitates the real conditions of haze scenes. 327 participants registered in the challenge and 23 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.