Abstract
Many proposals to generate shape models for tracking applications are based on a linear shape model, and a constraint that delimits the parameter values which generate feasible shapes. In this paper we introduce a novel approach to generate such models automatically. Given a training set, we determine the linear shape model as classical approaches, and model its associated constraint using a Gaussian Mixture Model, which is fully parameterized by a presented algorithm. Then, from this model we generate a collection of linear shape models of lower dimensionality, each one constrained by a single Gaussian model. This set of models represents better the training set, reducing the computational cost of tracking applications. To compare our proposal with the usual one, a comparison measure is defined, based on the Bayesian Information Criterion. Both modeling strategies are analyzed in a pedestrian tracking application, where our proposal claims to be more appropriate.
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.
Similar content being viewed by others
References
Y. Bar-Shalom, X. R. Li, and T. Kirubarajan. Estimation with Applications to Tracking and Navigation. John Wiley & Sons, 2001.
A.M. Baumberg. Learning Deformable Models for Tracking Human Motion. PhD thesis, The University of Leeds. School of Computer Studies, 1995.
A. Blake and A. Isard. Active Contours. Springer, 1998.
T. Cootes and C. Taylor. A mixture model for representing shape variation. Image Vision Computing, 1999.
T.F. Cootes and C. J. Taylor. Statistical models of appearance for computer vision. Technical report, Wolfson Image Analysis Unit. Imaging Science and Biomedical Engineering, University of Manchester, 2000.
T.F. Cootes, C.J. Taylor, D. Cooper, and J. Graham. Active shape models-their training and application. Computer Vision and Image Understanding, 1995.
N. M. Dempster, A.P. Laird and D.B. Rubin. Maximum likelihood from incomplete data via the EM algorithm. J. R. Statist. Soc. B, 39:185–197, 1977.
C. Fraley and A. Raftery. How many clusters? which clustering method? answers via model-based cluster analysis. Computer Journal, 41:578–588, 1998.
A. Heap and D. Hogg. Improving specificity in pdms using a hierarchical approach. In In Proc. British Machine Vision Conference, volume 1, pages 80–89, 1997.
T. Heap and D. Hogg. Wormholes in shape space: Tracking through discontinuous changes in shape. In Sixth International Conference on Computer Vision, pages 344–349, 1998.
M. Isard and A. Blake. Condensation-conditional density propagation for visual tracking. International Journal of Computer Vision, 1998.
R. Kalman. A new approach to linear filtering and prediction problems. Transactions of the ASME, 1960.
D. Ponsa and X. Roca. Unsupervised parameterisation of gaussian mixture models. Lecture Notes in Artificial Intelligence, to appear, 2002.
S. Richardson and P.J. Green. On Bayesian analysis of mixtures with an unknown number of components. Journal of the Royal Statistical Society (Series B), 59(4):731–758, 1997.
S. Roberts, C. Holmes, and D. Denison. Minimum entropy data partitioning using Reversible Jump Markov Chain Monte Carlo. IEEE Transactions on Pattern Analysis and Machine Intelligence, 23(8):909–915, August 2001.
S.J. Roberts, D. Husmeier, I. Rezek, and W. Penny. Bayesian Approaches To Gaussian Mixture Modelling. IEEE Transaction on Pattern Analysis and Machine Intelligence, 20(11):1133–1142, 1998.
Gideon Schwarz. Estimating the dimension of a model. Annals of Statistics, 6(2):461–464, March 1978.
Stephen P. Smith and Anil K. Jain. A test to determine the multivariate normality of a data set. IEEE Transactions on Pattern Analysis and Machine Intelligence, 10(5):757–761, September 1988.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ponsa, D., Roca, F.X. (2002). A Novel Approach to Generate Multiple Shape Models for Tracking Applications. In: Perales, F.J., Hancock, E.R. (eds) Articulated Motion and Deformable Objects. AMDO 2002. Lecture Notes in Computer Science, vol 2492. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-36138-3_7
Download citation
DOI: https://doi.org/10.1007/3-540-36138-3_7
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-00149-2
Online ISBN: 978-3-540-36138-1
eBook Packages: Springer Book Archive