Similar articles

Abstract: This novel method of pedestrian tracking using Support Vector (PTSV) proposed for a video surveillance instrument combines the Support Vector Machine (SVM) classifier into an optic-flow based tracker. The traditional method using optical flow tracks objects by minimizing an intensity difference function between successive frames, while PTSV tracks objects by maximizing the SVM classification score. As the SVM classifier for object and non-object is pre-trained, there is need only to classify an image block as object or non-object without having to compare the pixel region of the tracked object in the previous frame. To account for large motions between successive frames we build pyramids from the support vectors and use a coarse-to-fine scan in the classification stage. To accelerate the training of SVM, a Sequential Minimal Optimization Method (SMO) is adopted. The results of using a kernel-PTSV for pedestrian tracking from real time video are shown at the end. Comparative experimental results showed that PTSV improves the reliability of tracking compared to that of traditional tracking method using optical flow.

[1] Bergen, J.R., Anandan, P., Hanna, K., J. and Hingorani, R., 1993. Hierarchical Model-Based Motion Estimation. In: Motion Analysis and Image Sequence Processing. Sezan, M.I. and Lagendijk, R.L. (eds), Kluwer Academic Press, Dordrecht, Netherlands, p.257-232.

[2] Burges, C., 1996. Simplified Support Vector Decision Rules. Proceedings of the 13th International Conference on Machine Learning, San Mateo, Canada, p.71-77.

[3] Black, M. J. and Jepson, A., 1998. EigenTracking: robust matching and tracking of articulated objects using a view-based representation. International Journal of Computer Vision, 26(1):63-84.

[4] Gunn, S., 1998. Support Vector Machines for Classification and Regression. ISIS Technical Report ISIS-1-98, Image Speech Intelligent System Research Group, University of Southampton, England.

[5] Hui, H., Zhou, H. and Wang, L.Y., 2002. Optimal Gabor Filters Design for Fingerprint Recognition. Proceedings of SPIE, Annual Meeting 2002,Seattle, Washington, USA, 4790-85:351-356.

[6] Khan, S., Javed, O., Rasheed, Z. and Shah, M., 2001. Human Tracking in Multiple Cameras. The Eighth IEEE International Conference on Computer Vision, Vancouver, Canada, p.331-336.

[7] Liu, J.F. and Huang, D.R., 2001. Zerotrees and pyramidal lattice vector quantization for wavelet image coding. Journal of Image and Graphic, 6(A):229-232.

[8] Morik, K., Brockhausen, P. and Joachims, T., 1999. Combining Statistical Learning with a Knowledge-based Approach - A Case Study in Intensive Care Monitoring, Proc. 16th International Conf. on Machine Learning, Morgan Kaufman Publishers, San Mateo, Canada, p.268-277.

[9] Osuna, E., Freund, R. and Girosi, F., 1997. Training Support Vector Machines: An Application to Face Detection. Proc. of IEEE Conference on Computer Vision and Pattern Recognition, San Juan, Puerto Rico, p.130-136.

[10] Platt, J., 1999. Using Sparseness and Analytic QP to Speed Training of Support Sector Machines. In: Advances in Neural Information Processing System, M. S. Kearns, S. A. Solla, D. A. Cohn (eds), MIT Press, USA, 11:126-134.

[11] Simard, P., LeCun, Y. and Denker, J., 1993. Efficient Pattern Recognition using a New Transformation Distance. In: Advances in Neural Information Processing System, Lippmann, P.L., Moody, J.E., Touretzky, D.S (eds), Morgan Kaufman Publishers, San Mateo, CA, p.50-58.

[12] Scholkopf, B., Simard, P., Smola, A. and Vapnik, V., 1998. Prior Knowledge in Support Vector Kernels. In: Advances in Neural Information Processing Systems, M. I. Jordan, M. J. Kearns, S. A. Solla (eds), MIT Press, USA, 10:640-646.

[13] Vasconcelos, N. and Lippman, A., 1998. Multiresolution Tangent Distance for Affine-invariant Classification. In: Advances in Neural Information Processing Systems, M. I. Jordan, M. J. Kearns, S. A. Solla (eds), MIT Press, USA, 10:843-849.

[14] Wang, Z.Y., Chi, Z.R., Deng, D. and Cho, S.Y., 2001. Adaptive Processing of Tree-Structure Image Representation. IEEE Pacific Rim Conference on Multimedia, Beijing, China, p.989-995.

[15] Xin, D., Wu, Z.H. and Pan, Y.H., 2002. Probability output of multi-class support vector machines. Journal of Zhejiang University SCIENCE, 3(1):131-134.

[16] Zhou, H. and Wang, L.Y., 2001. Virtual instrument system software architecture description language. Journal of Zhejiang University SCIENCE, 2(4):411-415.