Henri Bouma , Jan Baan, Sander Borsboom, Kasper van Zon, Xinghan Luo, Ben Loke, Bram Stoeller, Hans van Kuilenburg, Judith Dijk , “WPSS: Watching people security services”, Proc. SPIE, Vol. 8901, (2013). http://dx.doi.org/10.1117/12.2031639 Copyright 2013 Society of Photo-Optical Instrumentation Engineers (SPIE). One print or electronic copy may be made for personal use only. Systematic reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited. WPSS: Watching people security services Henri Bouma 1* , Jan Baan 1 , Sander Borsboom 2 , Kasper van Zon 3 , Xinghan Luo 3 , Ben Loke 4 , Bram Stoeller 5 , Hans van Kuilenburg 3 , Judith Dijk 1 1 TNO, Oude Waalsdorperweg 63, 2597 AK The Hague, The Netherlands. 2 Cameramanager.com, Hogehilweg 19, 1101 CB Amsterdam, The Netherlands. 3 VicarVision, Singel 160, 1015 AH Amsterdam, The Netherlands. 4 Noldus Information Technology, Nieuwe Kanaal 5, 6709 PA Wageningen, The Netherlands. 5 Eagle Vision, Energiestraat 16B, 1411 AT Naarden, The Netherlands. ABSTRACT To improve security, the number of surveillance cameras is rapidly increasing. However, the number of human operators remains limited and only a selection of the video streams are observed. Intelligent software services can help to find people quickly, evaluate their behavior and show the most relevant and deviant patterns. We present a software platform that contributes to the retrieval and observation of humans and to the analysis of their behavior. The platform consists of mono- and stereo-camera tracking, re-identification, behavioral feature computation, track analysis, behavior interpretation and visualization. This system is demonstrated in a busy shopping mall with multiple cameras and different lighting conditions. Keywords: Surveillance, CCTV, security, tracking, behavior analysis, threat, biometry. 1. INTRODUCTION To improve security, the number of surveillance cameras is rapidly increasing. However, the number of human operators remains limited and only a selection of the video streams are observed. Intelligent software services can help to evaluate people’s behavior and show the most relevant and deviant patterns. In this paper, we present a software platform that contributes to the observation and analysis of human behavior. This platform was developed by a consortium of multiple companies in the WPSS-project. The platform consists of several components. It uses real-time tracking in multiple mono-cameras for overview, tracking in stereo-cameras for high location accuracy, and person re-identification to couple information between cameras. To enrich the person description, we used face analysis, body-part detection and 3D pose estimation. Finally, information is combined, tracks are analyzed, behavior is interpreted by a reasoning engine and directly visualized. The security services fuse all information and the results are made available in an online platform. This system has been tested and demonstrated in a crowded shopping mall with multiple cameras and different lighting conditions. The outline of the paper is as follows. The technical system is presented in Section 2. The demonstrator setup is described in Section 3 and the results are shown in Section 4. Finally, the conclusions are presented in Section 5. 2. METHOD The WPSS system consists of monocular and stereo cameras, video management, tracking and re-identification, measurement of behavioral features, track analysis and behavior analysis and graphical user interface to support interaction (Figure 1). Each is explained in more detail in the following subsections. * henri.bouma@tno.nl; phone +31 888 66 4054; http://www.tno.nl