Human detection and tracking using a Kinect camera for an autonomous service robot

This paper presents a novel method for people detection and tracking using depth images provided by Kinetic camera. The depth image captured by a Kinect camera is analysed using its histogram, allowing for the depth image to be divided in slices, making the retrieval of regions of interest a simple and computationally light process when compared to point clouds. These regions are then classified as human or not using a template matching technique. An efficient gradient descent algorithm is used to perform the template matching, using the RPROP algorithm, and the tracking is performed based on color image histogram comparison for each region of interest, in consecutive frames. The proposed method is viable for on-line detection and tracking of people and has been tested in a mobile platform in an unconstrained environment.

[1]  Martin Lauer,et al.  Calculating the Perfect Match: An Efficient and Accurate Approach for Robot Self-localization , 2005, RoboCup.

[2]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[3]  James W. Davis,et al.  A Two-Stage Template Approach to Person Detection in Thermal Imagery , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[4]  Jake K. Aggarwal,et al.  Human detection using depth information by Kinect , 2011, CVPR 2011 WORKSHOPS.

[5]  Javier Ruiz-del-Solar,et al.  Human Detection and Identification by Robots Using Thermal and Visual Information in Domestic Environments , 2011, Journal of Intelligent & Robotic Systems.

[6]  Gian Luca Foresti,et al.  Automatic detection and indexing of video-event shots for surveillance applications , 2002, IEEE Trans. Multim..

[7]  Hironobu Fujiyoshi,et al.  Real-Time Human Detection Using Relational Depth Similarity Features , 2010, ACCV.

[8]  Shuzhi Sam Ge,et al.  Robust Human Detection and Identification by Using Stereo and thermal Images in Human Robot Interaction , 2007, Int. J. Inf. Acquis..

[9]  Frederik Hegger,et al.  People Detection in 3d Point Clouds Using Local Surface Normals , 2012, RoboCup.

[10]  Larry S. Davis,et al.  W4: Real-Time Surveillance of People and Their Activities , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Kai Oliver Arras,et al.  People detection in RGB-D data , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.