Crowd Detection and Analysis for Surveillance Videos using Deep Learning
Aman Ahmed, Prateek Bansal, Atiya Khan, Neha Purohit
Abstract
Crowd identification and analysis has drawn a lot of attention recently, owing to a wide variety of video surveillance applications. We present a detailed review of crowd analysis and management, focusing on state-of-the-art methods for both controlled and unconstrained conditions. The paper illustrates both the advantages as well as disadvantages of state-of-the-art methods. Mass or crowd gathering can be seen at a lot of places like airports, sports stadiums, at various religious, educational, and entertainment-related events, etc. When tens of thousands of people gather in limited space, a tragedy is probably bound to happen. Automated video surveillance has become the need of the day and supports the analysis and management of data on a massive scale. It is very important to identify the presence of a crowd and detect the number of people in the gathering. This can prove very useful for the detection of sudden troupe build-up to avoid riots. Moreover, it can also be very useful in the Covid-19 pandemic situation to avoid people gathering at a place. This paper presents a system to detect the presence of a crowd by counting unique people and then performing crowd analysis. The crowd is analyzed by detecting the gender and age of people in the crowd.