Abstract
We propose an adaptive model update mechanism for mean-shift blob tracking. It is novel for us to use self-tuning Kalman filters for estimating object kernel-color distribution, i.e. kernel-histogram. Filtering residuals are employed for hypothesis testing in order to obtain a robust criterion for model update. Therefore, tracker has the ability to keep up with the changes of object appearance as well as the changes in scale. Moreover, over-update is avoided in the cases of severe occlusion and dramatic appearance changes. Various tracking sequences demonstrate the superior behavior of our tracker which runs in real-time with non-parameter initialization and is robust to appearance changes.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Cui, Y., Samarasekera, S., Huang, Q., Greiffenhagen, M.: Indoor monitoring via the collaboration between a peripheral sensor and a foveal sensor. IEEE Workshop on Visual Surveillance, 2–9 (1998)
Eleftheriadis, A., Jacquin, A.: Automatic Face Location Detection and Tracking for Model-Assisted Coding of Video Teleconference Sequences at Low Bit Rates. Signal Processing- Image Communication 3, 231–248 (1995)
Wactlar, H.D., Christel, M.G., Gong, Y., Hauptmann, A.G.: Lessons learned from the creation and deployment of a terabyte digital video library. IEEE Computer 2, 66–73 (1999)
Comaniciu, D., Ramesh, V., Meer, P.: Real-time tracking of non-rigid objects using mean shift. IEEE Int. Proc. Computer Vision and Pattern Recognition 2, 142–149 (2000)
Comaniciu, D., Ramesh, V., Meer, P.: Kernel-based object tracking. IEEE Trans. Pattern Analysis Machine Intelligence 5, 564–575 (2003)
Collins, R.T.: Mean shift blob tracking through scale space. IEEE Int. Proc. Computer Vision and Pattern Recognition 2, 234–240 (2003)
Nguyen, H.T., Worring, M., van den Boomagaard, R.: Occlusion robust adaptive template tracking. IEEE Int. Conf. Computer Vision 1, 678–683 (2001)
Blake, A., Curwen, R., Zisserman, A.: A framework for spatio-temporal control in the tracking of visual contour. Int. J. Computer Vision 2, 127–145 (1993)
Legters, G., Young, T.: A Mathematical Model for Computer Image Tracking. IEEE Trans. Pattern Analysis Machine Intelligence 6, 583–594 (1982)
Zhu, Z., Ji, Q., Fujimura, K.: Combining Kalman filtering and mean shift for real time eye tracking under active IR illumination. IEEE Int. Conf. Pattern Recognition 4, 318–321 (2002)
Fukanaga, K., Hostetler, L.D.: The estimation of the gradient of a density function, with applications in pattern recognition. IEEE Trans. Information Theory 1, 32–40 (1975)
Cheng, Y.: Mean shift, mode seeking and clustering. IEEE Trans. Pattern Analysis and Machine Intelligence 8, 790–799 (1995)
Maybeck, P.: Stochastic models, estimation and control. Academic Press, New York (1982)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Peng, NS., Yang, J. (2004). Mean-Shift Blob Tracking with Kernel-Color Distribution Estimate and Adaptive Model Update Criterion. In: Comaniciu, D., Mester, R., Kanatani, K., Suter, D. (eds) Statistical Methods in Video Processing. SMVP 2004. Lecture Notes in Computer Science, vol 3247. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30212-4_8
Download citation
DOI: https://doi.org/10.1007/978-3-540-30212-4_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23989-5
Online ISBN: 978-3-540-30212-4
eBook Packages: Springer Book Archive