Abstract
In this paper, we propose an adaptive Hamiltonian Monte Carlo sampling based tracking scheme within the Bayesian filtering framework. At the proposal step, the ordered over relaxation method is used to draw the momentum item for the joint state variable, which can suppress the random walk behavior. In addition, we design adaptive step-size based scheme to simulate the Hamiltonian dynamics in order to reduce the simulation error. The proposed method is compared with several state-of-the-art tracking algorithms. Extensive experimental results have shown its superiority in handling various types of abrupt motions compared to the traditional tracking algorithms.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Yang, H., Shao, L., Zheng, F., Wang, L., Song, Z.: Recent advances and trends in visual tracking: A review. Neurocomputing 74, 3823–3831 (2011)
Martinez-del-Rincon, J., Orrite, C., Medrano, C.: Rao-Blackwellised Particle Filter for Color-based Tracking. Pattern Recogn. Lett. 32, 210–220 (2011)
Isard, M., Blake, A.: CONDENSATION-Conditional Density Propagation for Visual Tacking. Int. J. Comput. Vision 29, 5–28 (1998)
Oron, S., Bar-Hillel, A., Levi, D., Avidan, S.: Locally Orderless Tracking. In: CVPR, pp. 1940–1947 (2012)
Khan, Z., Balch, T., Dellaert, F.: MCMC-based particle for tracking a variable number of interacting targets. IEEE T. Pattern Anal. 27, 1805–1819 (2005)
Lin, C., Wolf, W.: MCMC-based feature-guided particle filtering for tracking moving objects from a moving platform. In: International Conference on Computer Vision, pp. 1–2. IEEE Press, New York (2009)
Kwon, J., Lee, K.: Viusal tracking decomposition. In: International Conference on Computer Vision and Pattern Recognition, pp. 1–2. IEEE Press, New York (2010)
Kwon, J., Lee, K.M.: Tracking of abrupt motion using Wang-Landau Monte Carlo estimation. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008, Part I. LNCS, vol. 5302, pp. 387–400. Springer, Heidelberg (2008)
Kwon, J., Lee, K.: Wang-Landau Monte Carlo-based tracking methods for abrupt motions. IEEE T. Pattern Anal. 35, 1011–1024 (2013)
Zhou, X., Lu, Y., Lu, J., Zhou, J.: Abrupt motion tracking via intensively adaptive markov chain monte carlo sampling. IEEE T. Image Process. 21, 789–801 (2012)
Wang, F., Lu, M.: Efficient visual tracking via Hamiltonian Monte Carlo Markov chain. Comput. J. 59, 1102–1112 (2013)
Brooks, S., Gelman, A., Jones, G., Meng, X.: Handbook of Markov Chain Monte Carlo. Chapman and Hall/CRC press (2011)
Neal, R.M.: Suppressing random walks in Markov chain Monte Carlo using ordered overrexlation. Technical report, Department of Statistics, University of Toronto (1995)
Alfaki, M.: Improving efficiency in parameter estimation using the Hamiltonian Monte Carlo algorithm. Master’s Thesis, University of Bergen (2008)
Huang, W., Leimkuhler, B.: The adaptive Verlet method. SIAM J. Sci. Comput. 18, 239–256 (1997)
Holder, T., Leimkuhler, B., Reish, S.: Explicit variable step-size and time reversible integration. Appl. Numer. Math. 39, 367–377 (2001)
Roberts, O., Rosenthal, S.: Examples of adaptive MCMC. J. Comput. Graph. Stat. 18, 349–367 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Wang, F., Li, X., Lu, M., Xiao, Z. (2014). Robust Abrupt Motion Tracking via Adaptive Hamiltonian Monte Carlo Sampling. In: Pham, DN., Park, SB. (eds) PRICAI 2014: Trends in Artificial Intelligence. PRICAI 2014. Lecture Notes in Computer Science(), vol 8862. Springer, Cham. https://doi.org/10.1007/978-3-319-13560-1_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-13560-1_5
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-13559-5
Online ISBN: 978-3-319-13560-1
eBook Packages: Computer ScienceComputer Science (R0)