Implementation of social distancing for COVID 19 using YOLO

The contagious disease COVID 19 (Corona VIrus Disease 2019) which is caused by the novel CORONA virus has imposed lot of restrictions on the civilian movements within and between the countries. The WHO (World Health Organization) has passed a strict imposition to follow Social Distancing (SD) in order to prevent the spread of the disease. SD is nothing but keeping a least separation of 2 meters or 6 foot between the individuals in public places. Also, SD emphases on no contact or minimum contact between the people who are not from your household. The vaccination against COVID 19 disease might require a sufficient amount of time (few months to a year). Hence SD is the only current non pharmaceutical solution to curb the spread of this pandemic. On the other hand, the common people are not used to practice SD and maintain a safety bubble of 2m around them always. Hence there is a requirement of an automatic system that could help maintain the required SD without much difficulty. The object detection techniques are playing an important role in detecting various common objects that are seen in our day to day life. YOLO is the recent and real time object detection technology. In the field of computer vision, object identification is thought to be one of the most challenging tasks. In YOLO with a single evaluation, the bounding boxes and class probabilities are anticipated using only one neural network from the images directly. In the meantime, frame detection is treated as a regression problem. The intended scheme utilizes inverse perspective mapping and YOLOv3 based algorithm for detection of people accurately and to implement social distancing.