Efficient Mean Shift Clustering Using Exponential Integral Kernels

This paper presents a highly efficient algorithm for detecting and tracking humans and objects in video surveillance sequences. Mean shift clustering is applied on backgrounddifferenced image sequences. For efficiency, all calculations are performed on integral images. Novel corresponding exponential integral kernels are introduced to allow the application of nonuniform kernels for clustering, which dramatically increases robustness without giving up the efficiency of the integral data structures. Experimental results demonstrating the power of this approach are presented.

• S. Sutor
• R. Röhr
• G. Pujolle
• R. Reda

• Clustering
• Integral Images
• Kernels
• Person Detection
• Person Tracking
• Intelligent Video Surveillance.

[1] K. Fukunaga and L. Hostetler, "The Estimation of the Gradient of a
Density Function, with Applications in Pattern Recognition," IEEE
Transactions on Information Theory, vol. 21, pp. 32-40, 1975.
[2] C. Beleznai, B. Fr├╝hst├╝ck, H. Bischof, and W. Kropatsch, "Detecting
Humans in Groups Using a Fast Mean Shift Procedure", In Proceedings
of the 28th Workshop of the Austrian Association for Pattern
Recognition, vol. 179, pp. 71-78, 2004.
[3] P. Viola, M. Jones, "Rapid Object Detection using a Boosted Cascade of
Simple Features", IEEE Computer Society Conference on Computer
Vision and Pattern Recognition (CVPR'01) - Volume 1, pp. 511, 2001
[4] Y. Cheng, "Mean Shift, Mode Seeking and Clustering," IEEE
Transactions on Pattern Analysis and Machine Intelligence, vol. 17, pp.
790-799, 1995.
[5] F. Matusek, S. Sutor, K.Kruse, K. Kraus and R. Reda, Large-Scale
Video Surveillance Systems: New Performance Parameters and Metrics,
to be published at in Press, The Third International Conference on
Internet Monitoring and Protection, Bucharest June 29 - July 5, 2008
[6] A. Yilmaz, K. Shafique, and M. Shah, "Target Tracking in Airborne
Foward Looking Infrared Imagery," Image and Vision Computing
Journal, vol. 21, pp. 623-635, 2003.
[7] D. Comaniciu, V. Ramesch, and P. Meer, "Kernel-Based Object
Tracking," IEEE Transactions on Pattern Analysis and Machine
Intelligence, vol. 25, pp. 564-577, 2003.
[8] S. Sutor, "A Mean Shift Based Approach towards Automated Person
Tracking", Master Thesis, Vienna University of Technology, 2007