Abstract
We generalize the usual white noise acceleration target model by introducing an affine transformation to model the target aspect. This transformation is parameterized by scalar variables that describe the target shear, scale, and rotation and obey a first-order Markov chain. Our primary interest is in achieving robust Monte Carlo estimation in a complex state space under low signal-to-noise ratios, where a low-entropy unimodal likelihood function contributes to the failure of the conventional particle-filtering algorithms. Motivated by recently developed particle filter consistency checks, we develop a new track quality indicator that monitors tracking performance, triggering one of two actions as needed to improve the quality of the track. The first action is a boosting step, by which a local detector is defined based on the most recent tracker output to induce additional high-quality boosting particles. The original idea of boosting is extended here by encouraging positive interaction between the detector and the tracker. The second action is an adaptation step in which the system model self-adjusts to enhance tracking performance. In the context of affine-invariant target tracking, we compare these two techniques with respect to their effectiveness in improving the particle quality. We present experimental results that show that both techniques can improve the tracking performance by balancing the focus and the diversity of the particle distribution and by improving the particle filter consistency.
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.
Chapter's References
Gordon, N.J., Salmond, D.J., Smith, A.F.M. Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proceedings-F (Radar, Signal Process.) 140(2) (1993) 107– 113
Doucet, A., Freitas, J.F.G., Gordon, N.J.: Sequential Monte Carlo Methods in Practice. Springer-Verlag, New York (2001)
Heijden, F.V.D.: Consistency check for particle filters. IEEE Trans. Pattern Anal. Machine Intell. 28(1) (2006) 140–145
Chang, C., Ansari, R., Khokhar, A.: Multiple object tracking with kernel particle filter. In: Proc. IEEE Int. Conf. Comput. Vision, Pattern Recog. (2005) 566–573
Han, B., Zhu, Y., Comaniciu, D., Davis, L.: Kernel-based Bayesian filtering for object tracking. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recog. (2005) 227–234
Deutscher, J., Blake, A., Reid, I.D.: Articulated body motion capture by annealed particle filtering. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recog. (2000) 126–133
Bruno, M.G.S.: Sequential importance sampling filtering for target tracking in image sequences. IEEE Signal Process. Lett. 10(8) (2003) 246–249
Bruno, M.G.S., Moura, J.M.F.: Multiframe detection/tracking in clutter: Optimal performance. IEEE Trans. Aerosp. Electron. Syst. 37(3) (2001) 925–946
Zhou, S., Chellappa, R., Mogghaddam, B.: Adaptive visual tracking and recognition using appearance-adaptive models in particle filters. IEEE Trans. Image Process. 13(11) (2004) 1491–1505
Okuma, K., Taleghani, A., de Freitas, N., Little, J.J., Lowe, D.G.: A boosted particle filter: Multitarget detection and tracking. In: Proc. 8th Eur. Conf. Comput. Vision LNCS 3021 (2004) 28–39
Rathi, Y., Vaswani, N., Tannenbaum, A., Yezzi, A.: Particle filtering for geometric active contours with application to tracking moving and deforming objects. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recog. (2005) 2–9
Bruno, M.G.S.: Bayesian methods for multiaspect target tracking in image sequences. IEEE Trans. Signal Process. 52(7) (2004) 1848–1861
Wang, P., Rehg, J.M.: A modular approach to the analysis and evaluation of particle filters for figure tracking. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recog. (2006) 790–797
Moura, J.M.F., Balram, N.: Noncausal Gauss Markov random fields: Parameter structure and estimation. IEEE Trans. Inform. Theory 39(4) (1993) 1333–1355
Moura, J.M.F., Balram, N.: Recursive structure of noncousal Gauss Markov random fields. IEEE Trans. Inform. Theory 38(2) (1992) 334–354
Foley, J., van Dam, A., Feiner, S., Hughes, J.: Computer Graphics: Principles and Practice. 2nd edn. Addison-Wesley, Boston (1990)
Lichtenauer, J., Reinders, M., Hendriks, E.: Influence of the observation likelihood function on particle filtering performance in tracking applications. In: 6th IEEE Int. Conf. Automatic Face Gesture Recog. (2004) 767–772
Arulampalam, M.S., Maskell, S., Gordon, N., Clapp, T.: A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. IEEE Trans. Signal Process. 50(2) (2002) 174–188
Pitt, M.K., Shephard, N.: Filtering via simulation: auxiliary particle filters. J. Am. Stat. Assoc. 94(446) (1999) 590–599
Huang, Y., Djuric, P.M.: A hybrid importance function for particle filtering. IEEE Signal Process. Lett. 11(3) (2004) 404–406
Shen, C., Brooks, M.J., van den Hengel, A.: Augmented particle filtering for efficient visual tracking. In: Proc. IEEE Int. Conf. Image Process. 3 (2005) 856–859
Rui, Y., Chen, Y.: Better proposal distributions: Object tracking using unscented particle filter. In: Proc. IEEE Int. Conf. Comput. Vision, Pattern Recog. 2 (2001) 786–793
Wu, Y., Huang, T.S.: A co-inference approach to robust visual tracking. In: Proc. IEEE Int. Conf. Comput. Vision 2 (2001) 26–33
Chang, C., Ansari, R.: Kernel particle filter for visual tracking. IEEE Signal Process. Lett. 12(3) (2005) 242–245
Philomin, V., Duraiswami, R., Davis, L.S.: Quasi-random sampling for condensation. In: Proc. Eur. Conf. Comput. Vision 2 (2000) 134–149
Maggio, E., Cavallaro, A.: Hybrid particle filter and mean shift tracker with adaptive transition model. In: Proc. IEEE Int. Conf. Acoust. Speech Signal Proc., Philadelphia, (2005)
Torma, P., Szepesvri, C.: Enhancing particle filters using local likelihood sampling. In: Proc. 8th Eur. Conf. Comput. Vision, LNCS 3021 (2004) 16–27
Torma, P., Szepesvari, C.: On using likelihood-adjusted proposals in particle filtering: Local importance sampling. In: Proc. 4th Int. Symp. Image, Signal Process. Anal. (2005) 58–63
Andrieu, C., Doucet, A., Tadic, V.: Particle methods for change detection, system identification and control. Proc. IEEE 92(3) (2004) 423–438
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Fan, G., Venkataraman, V., Tang, L., Havlicek, J.P. (2009). On Boosted and Adaptive Particle Filters for Affine-Invariant Target Tracking in Infrared Imagery. In: Hammoud, R.I. (eds) Augmented Vision Perception in Infrared. Advances in Pattern Recognition. Springer, London. https://doi.org/10.1007/978-1-84800-277-7_18
Download citation
DOI: https://doi.org/10.1007/978-1-84800-277-7_18
Publisher Name: Springer, London
Print ISBN: 978-1-84800-276-0
Online ISBN: 978-1-84800-277-7
eBook Packages: Computer ScienceComputer Science (R0)