Advancing Image Understanding in Poor Visibility Environments: A Collective Benchmark Study

作者：Yang, Wenhan; Yuan, Ye; Ren, Wenqi; Liu, Jiaying; Scheirer, Walter J.; Wang, Zhangyang; Zhang, Taiheng; Zhong, Qiaoyong; Xie, Di; Pu, Shiliang; Zheng, Yuqiang; Qu, Yanyun; Xie, Yuhong; Chen, Liang; Li, Zhonghao; Hong, Chen; Jiang, Hao; Yang, Siyuan; Liu, Yan; Qu, Xiaochao; Wan, Pengfei; Zheng, Shuai; Zhong, Minhui; Su, Taiyi; He, Lingzhi; Guo, Yandong; Zhao, Yao; Zhu, Zhenfeng; Liang, Jinxiu; Wang, Jingwen; Chen, Tianyi; Quan, Yuhui; Xu, Yong; Liu, Bo; Liu, Xin; Sun, Qi; Lin, Tingyu

来源：IEEE Transactions on Image Processing, 2020, 29: 5737-5752.

DOI：10.1109/TIP.2020.2981922

• Poor visibility environment
• object detection
• face detection
• haze
• rain
• low-light conditions

Existing 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.

• 单位
  中国科学院研究生院; 北京交通大学; 天津大学; 厦门大学; 浙江大学; 同济大学; 国网电力科学研究院; 北京大学; 东北大学; 北京航空航天大学; 南阳理工学院