NTIRE 2018 Challenge on Image Dehazing: Methods and Results

• Main Authors: Ancuti, Cosmin, Zhao, Ruhao, Ma, Xiaoping, Qin, Yong, Jia, Limin, Friedel, Klaus, Ki, Sehwan, Sim, Hyeonjun, Choi, Jae-Seok, Kim, Sooye, Seo, Soomin, Ancuti, Codruta O., Kim, Saehun, Kim, Munchurl, Mondal, Ranjan, Santra, Sanchayan, Chanda, Bhabatosh, Liu, Jinlin, Mei, Kangfu, Li, Juncheng, Luyao, Fang, Faming, Timofte, Radu, Jiang, Aiwen, Qu, Xiaochao, Liu, Ting, Wang, Pengfei, Sun, Biao, Deng, Jiangfan, Zhao, Yuhang, Hong, Ming, Huang, Jingying, Chen, Yizhi, Van Gool, Luc, Chen, Erin, Yu, Xiaoli, Wu, Tingting, Genc, Anil, Engin, Deniz, Ekenel, Hazim Kemal, Liu, Wenzhe, Tong, Tong, Li, Gen, Gao, Qinquan, Zhang, Lei, Li, Zhan, Tang, Daofa, Chen, Yuling, Huo, Ziying, Alvarez-Gila, Aitor, Galdran, Adrian, Bria, Alessandro, Vazquez-Corral, Javier, Bertalmo, Marcelo, Demir, H. Seckin, Yang, Ming-Hsuan, Adil, Omer Faruk, Phung, Huynh Xuan, Jin, Xin, Chen, Jiale, Shan, Chaowei, Chen, Zhibo, Patel, Vishal M., Zhang, He, Sindagi, Vishwanath A.
• Format: Conference Proceeding
• Language: English
• Subjects: Cameras; Generators; Image color analysis; Image restoration; Lighting; Manuals; Meters
• ISSN: 2160-7516
• DOI: 10.1109/CVPRW.2018.00134

This paper reviews the first challenge on image dehazing (restoration of rich details in hazy image) with focus on proposed solutions and results. The challenge had 2 tracks. Track 1 employed the indoor images (using I-HAZE dataset), while Track 2 outdoor images (using O-HAZE dataset). The hazy images have been captured in presence of real haze, generated by professional haze machines. I-HAZE dataset contains 35 scenes that correspond to indoor domestic environments, with objects with different colors and specularities. O-HAZE contains 45 different outdoor scenes depicting the same visual content recorded in haze-free and hazy conditions, under the same illumination parameters. The dehazing process was learnable through provided pairs of haze-free and hazy train images. Each track had ~120 registered participants and 21 teams competed in the final testing phase. They gauge the state-of-the-art in image dehazing.