Kalman Filter Based Multiple Human Tracking Method in Auto-aligned Multi-Camera Environment

In this paper, we propose a method to automatically align multiple cameras and track multiple people observed in the multiple camera system. The proposed inter-camera aligning algorithm works based on rigid transformation. The tracking algorithm is based on the Kalman filtering method and merges several human positions through the iterative association and the process of model update. The proposed algorithms are verified by simulation and experimental data obtained in real environment.

[1]  Ramsey Michael Faragher,et al.  Understanding the Basis of the Kalman Filter Via a Simple and Intuitive Derivation [Lecture Notes] , 2012, IEEE Signal Processing Magazine.

[2]  J. Munkres.  ALGORITHMS FOR THE ASSIGNMENT AND TRANSIORTATION tROBLEMS* , 1957 .

[3]  Jordi Gonzàlez,et al.  A Simple Method of Multiple Camera Calibration for the Joint Top View Projection , 2008, Computer Recognition Systems 2.

[4]  Shaogang Gong,et al.  Tracking multiple people with a multi-camera system , 2001, Proceedings 2001 IEEE Workshop on Multi-Object Tracking.

[5]  Xiaogang Wang,et al.  Intelligent multi-camera video surveillance: A review , 2013, Pattern Recognit. Lett..

[6]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  J. Michael McCarthy,et al.  Introduction to theoretical kinematics , 1990 .

[8]  Rainer Stiefelhagen,et al.  Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics , 2008, EURASIP J. Image Video Process..

[9]  Ben J. A. Kröse,et al.  Multi-user identification and efficient user approaching by fusing robot and ambient sensors , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[10]  JongSuk Choi,et al.  Realistic 3D simulation of multiple human recognition over Perception Sensor Network , 2014, The 23rd IEEE International Symposium on Robot and Human Interactive Communication.