Minimum Tracking with SPSA and Applications to Image Registration

Oleg Granichin, Lev Gurevich, Alexander Vakhitov

2009

Abstract

An application of simultaneous perturbation stochastic approximation (SPSA) algorithm with two measurements per iteration to the problem of object tracking on video is discussed. The upper bound of mean square estimation error is determined in case of once differentiable functional and almost arbitrary noises. Weak restrictions on uncertainty allow to use random sampling instead of full pixelwise difference calculation. The experiments show significant increase in performance of object tracking comparing to the classical Lucas-Kanade algorithm. The results can be generalized to improve more recent kernel-based tracking methods.

References

  1. Lukas B., Kanade T., ”An Iterative Registration Technique with an Application to Stereo Vision,”. In Proc. Imaging Understanding Workshop., 1981, pp. 121-130.
  2. Tsin Y. Kernel Correlation as an Affinity Measure in Point-Sampled Vision Problems. PhD Thesis, Carnegie Mellon University, 2003.
  3. Popkov A. Yu., “Gradient methods for nonstationary unconstrained optimization problems,” Automat. Remote Control, No. 6, pp. 883-891, 2005.
  4. Polyak B. T., Introduction to Optimization. New York: Optimization Software. 1987.
  5. Kushner H., Yin G., Stochastic Approximation and Recursive Algorithms and Applications. Springer. 2003.
  6. Borkar V. S., Stochastic Approximation. A Dynamical Systems Viewpoint. Cambridge University Press. 2008.
  7. Guo L., “Stability of recursive stochastic tracking algorithms,” SIAM J. Control and Optimization, vol. 32, No 5, pp. 1195-1225, 1994.
  8. Granichin O. N., Polyak B. T., Randomized Algorithms of an Estimation and Optimization Under Almost Arbitrary Noises. Moscow: Nauka. 2003.
  9. Granichin O. N., “Procedure of stochastic approximation with disturbances at the input,” Automat. Remote Control, vol. 53, No 2, part 1, pp. 232-237, 1992.
  10. Granichin O. N., “A stochastic recursive procedure with dependent noises in the observation that uses sample perturbations in the input,” Vestnik Leningrad Univ. Math., vol. 22, No 1(4), pp. 27-31, 1989.
  11. Polyak B. T., Tsybakov A. B., “Optimal order of accuracy for search algorithms in stochastic optimization,” Problems of Information Transmission, vol. 26, No 2, pp. 126-133, 1990.
  12. Spall J. C., “Multivariate stochastic approximation using a simultaneous perturbation gradient approximation,” IEEE Trans. Automat. Contr., vol. 37, pp. 332-341, 1992.
  13. Spall J. C., “Developments in stochastic optimization algorithms with gradient approximations based on function measurements,” Proc. of the Winter Simulation Conf., pp. 207-214, 1994.
Download


Paper Citation


in Harvard Style

Granichin O., Gurevich L. and Vakhitov A. (2009). Minimum Tracking with SPSA and Applications to Image Registration . In Proceedings of the International Workshop on Networked embedded and control system technologies: European and Russian R&D cooperation - Volume 1: Workshop NESTER, (ICINCO 2009) ISBN 978-989-674-004-7, pages 66-74. DOI: 10.5220/0002271300660074


in Bibtex Style

@conference{workshop nester09,
author={Oleg Granichin and Lev Gurevich and Alexander Vakhitov},
title={Minimum Tracking with SPSA and Applications to Image Registration},
booktitle={Proceedings of the International Workshop on Networked embedded and control system technologies: European and Russian R&D cooperation - Volume 1: Workshop NESTER, (ICINCO 2009)},
year={2009},
pages={66-74},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002271300660074},
isbn={978-989-674-004-7},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Workshop on Networked embedded and control system technologies: European and Russian R&D cooperation - Volume 1: Workshop NESTER, (ICINCO 2009)
TI - Minimum Tracking with SPSA and Applications to Image Registration
SN - 978-989-674-004-7
AU - Granichin O.
AU - Gurevich L.
AU - Vakhitov A.
PY - 2009
SP - 66
EP - 74
DO - 10.5220/0002271300660074