Mobile User Identification by Camera-Based Motion Capture and Mobile Device Acceleration Sensors

Context-awareness using camera images is a promising technique for enabling ubiquitous computing and networking; however, it is still an open issue to identify mobile users, i.e., identifying an actual user with a mobile device from people in an area. This paper discusses a mobile user identification method mapping users in the camera images to mobile devices connected to an access point. The proposed scheme focuses on acceleration of a hand-held mobile device and that of a mobile user's hand, which synchronously vary when the mobile user utilizes the device. The scheme obtains the user's hand motion from cameras by motion capture, converts that data into acceleration of the user's hand, calculates the correlations between the value of the acceleration of the user's hands and devices, and solves a matching problem. Experimental results show that the proposed scheme identifies mobile users with 100% accuracy when users walk at 1 m/s or when users walk at 0.5 m/s and stop to use their mobile devices. The proposed scheme also identifies with greater than 94% accuracy even when the numbers of users and mobile devices are different.

[1]  Alex Pentland,et al.  Pfinder: real-time tracking of the human body , 1996, Proceedings of the Second International Conference on Automatic Face and Gesture Recognition.

[2]  Masahiro Morikura,et al.  Proactive Base Station Selection Based on Human Blockage Prediction Using RGB-D Cameras for mmWave Communications , 2014, 2015 IEEE Global Communications Conference (GLOBECOM).

[3]  Masahiro Morikura,et al.  Recurrent neural network-based received signal strength estimation using depth images for mmWave communications , 2018, 2018 15th IEEE Annual Consumer Communications & Networking Conference (CCNC).

[4]  Bernt Schiele,et al.  Monocular 3D pose estimation and tracking by detection , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[5]  Andreas Savvides,et al.  Towards Cooperative Localization of Wearable Sensors using Accelerometers and Cameras , 2010, 2010 Proceedings IEEE INFOCOM.

[6]  Alex Pentland,et al.  Pfinder: Real-Time Tracking of the Human Body , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Alberto Del Bimbo,et al.  Soccer players identification based on visual local features , 2007, CIVR '07.

[8]  Shuping Dang,et al.  Hybrid forecasting model of power demand based on three-stage synthesis and stochastically self-adapting mechanism , 2014, 2014 IEEE International Energy Conference (ENERGYCON).

[9]  Teruo Higashino,et al.  Player identification by motion features in sport videos using wearable sensors , 2016, 2016 Ninth International Conference on Mobile Computing and Ubiquitous Networking (ICMU).

[10]  Ioannis A. Kakadiaris,et al.  A Review of Human Activity Recognition Methods , 2015, Front. Robot. AI.

[11]  Didier Dubois,et al.  Gradual inference rules in approximate reasoning , 1992, Inf. Sci..