Strategic Way to Count the Number of People in a Room Using Multiple Kinect Cameras Yuchen Liu, Sourav Chakraborty, Ashutosh Kumar, Rakesh Seal and Sohang Sengupta Abstract This paper proposes a system in which we use multiple Kinect cameras placed in very strategic positions to accurately detect the presence of the number of human beings present in a room and also keep a count of the number entering and leaving the room. Banking physical security and enterprise data centre security are some of the areas where such technologies can effectively make a difference in terms of physical security practices and attendance. Keywords Kinect camera · Face detection · OpenCV · Overhead camera · Deep learning · Facenet · Computer vision · Convolutional neural network (CNN) Y. Liu Department of Electrical and Computer Engineering, Boston University, Boston, USA e-mail: lyc@bu.edu S. Chakraborty Imperial College London Business School, London, UK e-mail: sourav.chakraborty18@imperial.ac.uk A. Kumar Department of Information Technology, Institute of Engineering and Management, Kolkata, India e-mail: iashu12966@gmail.com R. Seal (B ) Department of Electronics and Communication Engineering, Institute of Engineering and Management, Kolkata, India e-mail: rakeshseal0@gmail.com S. Sengupta Department of Computer Science and Engineering, Institute of Engineering and Management, Kolkata, India e-mail: mailsohang@gmail.com © Springer Nature Singapore Pte Ltd. 2020 M. Chakraborty et al. (eds.), Proceedings of International Ethical Hacking Conference 2019, Advances in Intelligent Systems and Computing 1065, https://doi.org/10.1007/978-981-15-0361-0_3 39