Advancing Image Understanding in Poor Visibility Environments: A Collective Benchmark Study
Yang, Wenhan1; Yuan, Ye2; Ren, Wenqi3; Liu, Jiaying1; Scheirer, Walter J.4; Wang, Zhangyang2; Zhang, Taiheng5; Zhong, Qiaoyong6; Xie, Di6; Pu, Shiliang6; Zheng, Yuqiang7; Qu, Yanyun7; Xie, Yuhong7; Chen, Liang7; Li, Zhonghao7; Hong, Chen7; Jiang, Hao8; Yang, Siyuan8; Liu, Yan8; Qu, Xiaochao8; Wan, Pengfei8; Zheng, Shuai9; Zhong, Minhui9; Su, Taiyi10; He, Lingzhi9; Guo, Yandong11; Zhao, Yao9; Zhu, Zhenfeng9; Liang, Jinxiu12; Wang, Jingwen13; Chen, Tianyi12; Quan, Yuhui12; Xu, Yong12; Liu, Bo14; Liu, Xin14; Sun, Qi14; Lin, Tingyu14; Li, Xiaochuan14; Lu, Feng14; Gu, Lin15; Zhou, Shengdi16; Cao, Cong16; Zhang, Shifeng17; Chi, Cheng18; Zhuang, Chubing17; Lei, Zhen17; Li, Stan Z.19,20; Wang, Shizheng21,22; Liu, Ruizhe22,23; Yi, Dong24; Zuo, Zheming25; Chi, Jianning26; Wang, Huan26; Wang, Kai18; Liu, Yixiu26; Gao, Xingyu21; Chen, Zhenyu27,28; Guo, Chang29; Li, Yongzhou29; Zhong, Huicai; Huang, Jing30; Guo, Heng30; Yang, Jianfei30; Liao, Wenjuan31; Yang, Jiangang18; Zhou, Liguo32; Feng, Mingyue32; Qin, Likun18
Corresponding AuthorScheirer, Walter J.( ; Wang, Zhangyang(
AbstractExisting enhancement methods are empirically expected to help the high-level end computer vision task: however, that is observed to not always be the case in practice. We focus on object or face detection in poor visibility enhancements caused by bad weathers (haze, rain) and low light conditions. To provide a more thorough examination and fair comparison, we introduce three benchmark sets collected in real-world hazy, rainy, and low-light conditions, respectively, with annotated objects/faces. We launched the UG(2+) challenge Track 2 competition in IEEE CVPR 2019, aiming to evoke a comprehensive discussion and exploration about whether and how low-level vision techniques can benefit the high-level automatic visual recognition in various scenarios. To our best knowledge, this is the first and currently largest effort of its kind. Baseline results by cascading existing enhancement and detection models are reported, indicating the highly challenging nature of our new data as well as the large room for further technical innovations. Thanks to a large participation from the research community, we are able to analyze representative team solutions, striving to better identify the strengths and limitations of existing mindsets as well as the future directions.
KeywordPoor visibility environment object detection face detection haze rain low-light conditions
Indexed BySCI
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000529943000015
Citation statistics
Document Type期刊论文
Corresponding AuthorScheirer, Walter J.; Wang, Zhangyang
Affiliation1.Peking Univ, Wangxuan Inst Comp Technol, Beijing 100080, Peoples R China
2.Texas A&M Univ, Dept Comp Sci & Engn, College Stn, TX USA
3.Chinese Acad Sci, Inst Informat Engn, State Key Lab Informat Secur, Beijing 100864, Peoples R China
4.Univ Notre Dame, Dept Comp Sci & Engn, Notre Dame, IN 46556 USA
5.Zhejiang Univ, Dept Mech Engn, Hangzhou 310027, Peoples R China
6.Hikvis Res Inst, Hangzhou 310051, Peoples R China
7.Xiamen Univ, Sch Informat, Fujian Key Lab Sensing & Comp Smart City, Xiamen 361001, Peoples R China
8.Meitu Inc, Mtlab, Beijing 100080, Peoples R China
9.Beijing Jiaotong Univ, Inst Informat Sci, Beijing 100044, Peoples R China
10.Tongji Univ, Dept Comp Sci & Technol, Shanghai 201804, Peoples R China
11.Xpeng Motors, Beijing 100080, Peoples R China
12.South China Univ Technol, Sch Comp Sci & Engn, Guangzhou 510006, Peoples R China
13.Tencent AI Lab, Shenzhen 518000, Peoples R China
14.Beihang Univ, Sch Comp Sci & Engn, Beijing 100191, Peoples R China
15.RIKEN AIP, Tokyo 1030027, Japan
16.Tianjin Univ, Sch Elect & Informat Engn, Tianjin 300072, Peoples R China
17.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
18.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
19.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
20.Westlake Univ, Sch Engn, Hangzhou 310024, Peoples R China
21.Chinese Acad Sci, Inst Microelect, Beijing 100029, Peoples R China
22.Chinese Acad Sci, R&D Ctr Internet Things, Wuxi 214200, Jiangsu, Peoples R China
23.Sunway AI Co Ltd, Zhuhai 519000, Peoples R China
24.Winsense Inc, Beijing 100080, Peoples R China
25.Univ Durham, Dept Comp Sci, Durham 46390, England
26.Northeastern Univ, Shenyang 110819, Peoples R China
27.State Grid Corp China, Big Data Ctr, Beijing 100031, Peoples R China
28.China Elect Power Res Inst, Beijing 100031, Peoples R China
29.Chinese Acad Sci, Inst Microelect, Beijing 100190, Peoples R China
30.Nanyang Technol Univ, Singapore 639798, Singapore
31.Australian Natl Univ, Canberra, ACT 0200, Australia
32.Tech Univ Munich, Dept Informat, Garching 85748, Germany
Recommended Citation
GB/T 7714
Yang, Wenhan,Yuan, Ye,Ren, Wenqi,et al. Advancing Image Understanding in Poor Visibility Environments: A Collective Benchmark Study[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2020,29:5737-5752.
APA Yang, Wenhan.,Yuan, Ye.,Ren, Wenqi.,Liu, Jiaying.,Scheirer, Walter J..,...&Qin, Likun.(2020).Advancing Image Understanding in Poor Visibility Environments: A Collective Benchmark Study.IEEE TRANSACTIONS ON IMAGE PROCESSING,29,5737-5752.
MLA Yang, Wenhan,et al."Advancing Image Understanding in Poor Visibility Environments: A Collective Benchmark Study".IEEE TRANSACTIONS ON IMAGE PROCESSING 29(2020):5737-5752.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Yang, Wenhan]'s Articles
[Yuan, Ye]'s Articles
[Ren, Wenqi]'s Articles
Baidu academic
Similar articles in Baidu academic
[Yang, Wenhan]'s Articles
[Yuan, Ye]'s Articles
[Ren, Wenqi]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Yang, Wenhan]'s Articles
[Yuan, Ye]'s Articles
[Ren, Wenqi]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.