Tracking mean shift clustered point clouds for 3D surveillance

We present in this paper a method of tracking multiple objects (people) in 3D for application in video surveillance. The tracking method is designed to work on images with objects at low resolution and has two major contributions. First we propose a way to generate 3D point clouds that imposes multiple constraints (both geometric and appearance-based) to ensure minimal noise in the 3D data. Second, we incorporate a method to group the points into clouds (or clusters) that correspond to objects in the environment being imaged. We show that this method is more powerful than current 3D tracking techniques that try to fuse 2D tracking information into 3D tracks. A comparison to competing 3D tracking methods are shown, and performance and limitations are discussed.

[1]  Zhengyou Zhang.  A New Multistage Approach to Motion and Structure Estimation by Gradually Enforcing Geometric Constraints , 1998, ACCV.

[2]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[3]  Gregory D. Hager,et al.  A Particle Filter without Dynamics for Robust 3D Face Tracking , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[4]  Tim J. Ellis,et al.  Multi camera image tracking , 2006, Image Vis. Comput..

[5]  Dorin Comaniciu,et al.  Kernel-Based Object Tracking , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[7]  Roger Y. Tsai,et al.  A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses , 1987, IEEE J. Robotics Autom..

[8]  Ian D. Reid,et al.  The Active Recovery of 3D Motion Trajectories and Their Use in Prediction , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Frank Dellaert,et al.  Robust car tracking using Kalman filtering and Bayesian templates , 1998, Other Conferences.

[10]  Dorin Comaniciu,et al.  Real-time tracking of non-rigid objects using mean shift , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[11]  Michael Isard,et al.  CONDENSATION—Conditional Density Propagation for Visual Tracking , 1998, International Journal of Computer Vision.

[12]  Rachid Deriche,et al.  A Robust Technique for Matching two Uncalibrated Images Through the Recovery of the Unknown Epipolar Geometry , 1995, Artif. Intell..

[13]  Jake K. Aggarwal,et al.  Object tracking in an outdoor environment using fusion of features and cameras , 2006, Image Vis. Comput..

[14]  Osama Masoud,et al.  A novel method for tracking and counting pedestrians in real-time using a single camera , 2001, IEEE Trans. Veh. Technol..

[15]  Christopher O. Jaynes,et al.  Moving Shadow Detection using a Combined Geometric and Color Classification Approach , 2005, 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1.

[16]  Maarten Vergauwen,et al.  Wide-baseline 3D reconstruction from digital stills , 2003 .

[17]  Mubarak Shah,et al.  A Multiview Approach to Tracking People in Crowded Scenes Using a Planar Homography Constraint , 2006, ECCV.

[18]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[19]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[20]  Stan Sclaroff,et al.  Improved Tracking of Multiple Humans with Trajectory Predcition and Occlusion Modeling , 1998 .