Bird's-eye View Social Distancing Analysis System
Bird's-eye View Social Distancing Analysis System
复制标题
DOI：
10.1109/iccworkshops53468.2022.9814627
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Zhengye Yang;Mingfei Sun-;Hongzhe Ye;Zihao Xiong;G. Zussman;Z. Kostić
中科院分区：
文献类型：
--
作者：
Zhengye Yang;Mingfei Sun-;Hongzhe Ye;Zihao Xiong;G. Zussman;Z. Kostić
保持社交距离可以降低COVID-19等呼吸道大流行病的感染率。在大都市中，交通路口特别适合监测和评估社交距离行为。因此，在本文中，我们提出并评估了一种实时保护隐私的社会距离分析系统（B-SDA），该系统使用鸟瞰视频记录行人穿过交通路口。我们设计了视频预处理、目标检测和跟踪的算法，这些算法植根于已知的计算机视觉和深度学习技术，但经过修改以解决检测由高架摄像机捕获的非常小的物体/行人的问题。我们提出了一种结合行人分组的方法来检测社交距离违规行为，该方法的F1得分为0.92。利用B-SDA比较大流行前和大流行期间曼哈顿上城区行人的行为，结果显示，大流行期间违反社交距离的比例为15.6%，明显低于大流行前基线的31.4%。
Social distancing can reduce the infection rates in respiratory pandemics such as COVID-19. Traffic intersections are particularly suitable for monitoring and evaluation of social distancing behavior in metropolises. Hence, in this paper, we propose and evaluate a real-time privacy-preserving social distancing analysis system (B-SDA), which uses bird's-eye view video recordings of pedestrians who cross traffic intersections. We devise algorithms for video pre-processing, object detection, and tracking which are rooted in the known computer-vision and deep learning techniques, but modified to address the problem of detecting very small objects/pedestrians captured by a highly elevated camera. We propose a method for incorporating pedestrian grouping for detection of social distancing violations, which achieves 0.92 F1 score. B-SDA is used to compare pedestrian behavior in pre-pandemic and during-pandemic videos in uptown Manhattan, showing that the social distancing violation rate of 15.6% during the pandemic is notably lower than 31.4% prenandemic baseline.