[论文解读] A Vision-based Social Distancing and Critical Density Detection System for COVID-19

Dongfang Yang, Ekim Yurtsever+3 more|arXiv (Cornell University)|Jul 7, 2020

COVID-19 epidemiological studies参考文献 39被引用 40

一句话总结

本文提出一个实时、隐私保护的视觉系统，使用单摄像头检测社交距离违规，并通过定义的关键密度来调控人群流入，附带开源代码。

ABSTRACT

Social distancing has been proven as an effective measure against the spread of the infectious COronaVIrus Disease 2019 (COVID-19). However, individuals are not used to tracking the required 6-feet (2-meters) distance between themselves and their surroundings. An active surveillance system capable of detecting distances between individuals and warning them can slow down the spread of the deadly disease. Furthermore, measuring social density in a region of interest (ROI) and modulating inflow can decrease social distancing violation occurrence chance. On the other hand, recording data and labeling individuals who do not follow the measures will breach individuals' rights in free-societies. Here we propose an Artificial Intelligence (AI) based real-time social distancing detection and warning system considering four important ethical factors: (1) the system should never record/cache data, (2) the warnings should not target the individuals, (3) no human supervisor should be in the detection/warning loop, and (4) the code should be open-source and accessible to the public. Against this backdrop, we propose using a monocular camera and deep learning-based real-time object detectors to measure social distancing. If a violation is detected, a non-intrusive audio-visual warning signal is emitted without targeting the individual who breached the social distancing measure. Also, if the social density is over a critical value, the system sends a control signal to modulate inflow into the ROI. We tested the proposed method across real-world datasets to measure its generality and performance. The proposed method is ready for deployment, and our code is open-sourced.

研究动机与目标

• 激发并解决社交距离监控中的隐私和非侵入性关注点。
• 开发一个实时的基于AI的系统，在不记录数据的情况下检测人际距离。
• 引入一种新颖的关键社交通密度指标以调控流入并防止拥挤。
• 提供在真实人群数据集上得到验证的开源解决方案。

提出的方法

• 使用预训练的深度卷积神经网络检测器在图像空间检测行人。
• 通过反单应变换将图像坐标映射到真实世界的鸟瞰视图(BEV)，以获得地面平面位置。
• 计算行人之间的距离并按阈值 d_c 计数违规。
• 使用违规数 v 相对于密度 rho 的线性回归来定义并估计关键社交通密度 rho_c，在 v=0 时采用 95%预测区间。
• 在违规发生时发出非侵入式音视频警告，并在 rho > rho_c 时调节感兴趣区域ROI的流入。

实验结果

研究问题

• RQ1 在不记录数据的前提下，使用固定单目摄像头及AI检测器是否能实时准确测量行人之间的距离？
• RQ2 如何定义并估计一个关键社交通密度，以将社交距离违规保持在目标概率以下？
• RQ3 预训练检测器在城市、室内以及类似地铁的人群场景中用于社交距离监控时，能否良好泛化？

主要发现

• Faster R-CNN 和 YOLOv4 在所测试的数据集上实现了实时行人检测。
• 通过反单应变换的 BEV 映射实现了在真实世界坐标中测量行人之间的距离。
• 社交通密度 rho 与违规数量 v 正相关，从而实现基于线性回归的 rho_c 估计。
• 所计算的关键密度 rho_c 在 Oxford Town Center、Mall 和 Train Station 数据集之间相似。
• 系统支持非侵入式警告和流入调节，且不记录数据或针对个人。