Social Distancing Monitoring for Real-Time Deep Learning Framework

In this post-pandemic era where and when another COVID-19 wave will hit, it is essential to maintain social distance. Avoiding large public gatherings and keeping the public spaces monitored are a reliable measure against the spread of virus, but maintaining this is not easy. Social distancing is an effective measure against the Coronavirus Disease. However, it is not possible for the general mass to imagine a safety bubble around them. According to the World Health Organization, individuals are required to maintain 6-feet (2-m) distance between themselves and their surroundings. In this paper, the prototype model for social distance detection using deep learning to detect and monitor the individual is described. A real-time surveillance system is proposed with non-alarming visual cue as a solution to the above-mentioned problem. The pandemic is going to remain an impending threat to the world and through proposed method, to take precautionary measures for another COVID-19 outbreak. The model is recorded with an accuracy of 98.6% on real-time dataset.