English

NTIRE 2020 Challenge on NonHomogeneous Dehazing

Computer Vision and Pattern Recognition 2020-05-08 v1

Abstract

This paper reviews the NTIRE 2020 Challenge on NonHomogeneous Dehazing of images (restoration of rich details in hazy image). We focus on the proposed solutions and their results evaluated on NH-Haze, a novel dataset consisting of 55 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. NH-Haze is the first realistic nonhomogeneous haze dataset that provides ground truth images. The nonhomogeneous haze has been produced using a professional haze generator that imitates the real conditions of haze scenes. 168 participants registered in the challenge and 27 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.

Keywords

Cite

@article{arxiv.2005.03457,
  title  = {NTIRE 2020 Challenge on NonHomogeneous Dehazing},
  author = {Codruta O. Ancuti and Cosmin Ancuti and Florin-Alexandru Vasluianu and Radu Timofte and Jing Liu and Haiyan Wu and Yuan Xie and Yanyun Qu and Lizhuang Ma and Ziling Huang and Qili Deng and Ju-Chin Chao and Tsung-Shan Yang and Peng-Wen Chen and Po-Min Hsu and Tzu-Yi Liao and Chung-En Sun and Pei-Yuan Wu and Jeonghyeok Do and Jongmin Park and Munchurl Kim and Kareem Metwaly and Xuelu Li and Tiantong Guo and Vishal Monga and Mingzhao Yu and Venkateswararao Cherukuri and Shiue-Yuan Chuang and Tsung-Nan Lin and David Lee and Jerome Chang and Zhan-Han Wang and Yu-Bang Chang and Chang-Hong Lin and Yu Dong and Hongyu Zhou and Xiangzhen Kong and Sourya Dipta Das and Saikat Dutta and Xuan Zhao and Bing Ouyang and Dennis Estrada and Meiqi Wang and Tianqi Su and Siyi Chen and Bangyong Sun and Vincent Whannou de Dravo and Zhe Yu and Pratik Narang and Aryan Mehra and Navaneeth Raghunath and Murari Mandal},
  journal= {arXiv preprint arXiv:2005.03457},
  year   = {2020}
}

Comments

CVPR Workshops Proceedings 2020