Real Time Social Distance Monitoring with Alarm System

World Health Organization (WHO) recommends social distancing as one of the most effective measures to prevent the spread of COVID-19 infection. Ensuring strict adherence to the norms is challenging especially in public places and work environments. Hence, we propose a system using the existing CCTV cameras or an IP camera and develop a Computer Vision and YOLOv3-based Deep Neural Network (DNN) model for automatic people detection. The model works in both in both indoor and outdoor environments. The model is trained using Microsoft Common Objects (COCO dataset). The model developed can alternatively be deployed for other applications including detection of autonomous cars, anomaly detection, crowd analysis etc.