Published Paper Details:

The outbreak of the COVID-19 pandemic has resulted in a widespread need for social distancing in order to slow the spread of the disease. In many cases, this has meant the closure of public spaces and the implementation of strict rules regarding the maximum number of people who can be in a given space at one time. However, enforcing these rules can be difficult, particularly in large spaces. In this paper, we propose a system for automatically monitoring social distancing in public spaces using a combination of person detection and tracking. Our system is based on the YOLO v3 object detection algorithm, which we fine-tune for better performance on person detection. We then use the deep sort algorithm to track the detected people and ensure that they are complying with social distancing rules. Our system is capable of monitoring multiple people simultaneously and can provide alerts if people are not following the rules. Overall, our system provides a simple and effective way to monitor social distancing in public places. It is accurate, scalable, and can be easily deployed in a wide range of settings. Our system achieves an accuracy of 97.5% for person detection and 95.4% for social distancing compliance.