Abstract
A Point Distribution Model requires first the choice of an appropriate representation for the data and then the estimation of the density within this representation. Independent Component Analysis is a linear transform that represents the data in a space where statistical dependencies between the components are minimized. In this paper, we propose Independent Component Analysis as a representation for point distributions. We observe that within this representation, the density estimation is greatly simplified and propose solutions to the most common problems concerning shapes. Mainly, testing shape feasibility and finding the nearest feasible shape. We also observe how the description of shape deformations in terms of statistically independent modes provides a more intuitive and manageable framework. We perform experiments to illustrate the results and compare them with existing approaches.
This work is supported by CICYT and EU grants TAP98-0631 and 2FD97-0220 and the Secretaría de Estado de Educación, Universidades, Investigación y Desarollo from the Ministerio de Educación y Cultura de España.
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
A. Baumberg and D. Hogg. An efficient method for contour tracking using active shape models. Technical report, University of Leeds, School of Computer Studies, November 1994.
A. Bell and T. Sejnowski. An information-maximization approach for blind signal separation. Neural Computation, 7:1129–1159, 1995.
A. Bell and T. Sejnowski. The ‘independent components’ of natural scenes are edge filters. Neural Computation, 11:1739–1768, 1999.
C. Bishop. Neural Networks for Pattern Recognition. Oxford University Press, 1997.
R. Bowden, T. Mitchell, and M. Sahardi. Cluster based non-linear principle component analysis. IEE Electronic Letters, 33(22):1858–1858, 1997.
J. Cardoso. Infomax and maximum likelihood for source separation. IEEE Letters on Signal Processing, 4:112–114, 1997.
P. Comon. Independent component analysis-a new concept? Signal Processing, 36:287–314, 1994.
T. Cootes and C. Taylor. A mixture model for representing shape variation. In Clark A.F., ed. British Machine Vision Conference 1997, BMVC’97, volume 1, pages 110–119. University of Essex, UK:BMVA, 1997.
T. Cootes, C. Taylor, D. Cooper, and J. Graham. Active shape models-their training and application. Computer Vision and Image Understanding, 61:38–59, 1995.
A. Dempster, N. Laird, and D. Rubin. Maximum likelihood for incomplete data via the em algorithm. Journal of the Royal Statistical Society, 39:1–38, 1977.
J. Gower. Generalized procrustes analysis. Psychometrika, 40:33–51, 1975.
A. Heap and D. Hogg. 3d deformable hand models. In Gesture Workshop, York, UK, March 1996.
A. Heap and D. Hogg. Improving speci_city in pdms using a hierarchical approach. In Clark A.F., ed. British Machine Vision Conference 1997, BMVC’97, volume 1, pages 80–89. University of Essex, UK:BMVA, 1997.
A. Hyvärinen. Survey on independent component analysis. Neural Computing Surveys, 2:94–128, 1999.
A. Hyvärinen and E. Oja. A fast fixed-point algorithm for independent component analysis. Neural Computation, 9:1483–1492, 1999.
T. Lee, M. Girolami, and T. Sejnowski. Independent component analysis using an extended infomax algorithm for mixed sub-gaussian and super-gaussian sources. Neural Computation, 11:609–633, 1998.
D. Pham, P. Garrat, and C. Jutten. Separation of a mixture of independent sources through a maximum likelihood approach. In EUSIPCO, pages 771–774, 1992.
J. Porrill and J. Stone. Undercomplete independent component analysis for signal separation and dimension reduction. Technical report, The University of Sheffeld, Department of Psychology, 1998.
B. Silverman. Density Estimation. Chapman and Hall, 1986.
P. Sozou, T. Cootes, C. Taylor, and E. Di-Mauro. A non-linear generalization of pdms using polynomial regression. In Hancock E., ed. British Machine Vision Conference 1994, BMVC’94, volume 1, pages 397–406. University of York, UK:BMVA, 1994.
P. Sozou, T. Cootes, C. Taylor, and E. Di-Mauro. Non-linear point distribution modeling using a multi-layer perceptron. In Pycock D., ed. British Machine Vision Conference 1995, BMVC’95, volume 1, pages 107–116. University of Birmingham, UK:BMVA, 1995.
M. Stegmann. On properties of active shape models. Technical report, Technical University of Denmark, Department of Mathematical Modeling, March 2000.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bressan, M., Vitrià, J. (2001). Independent Modes of Variation in Point Distribution Models. In: Arcelli, C., Cordella, L.P., di Baja, G.S. (eds) Visual Form 2001. IWVF 2001. Lecture Notes in Computer Science, vol 2059. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45129-3_10
Download citation
DOI: https://doi.org/10.1007/3-540-45129-3_10
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42120-7
Online ISBN: 978-3-540-45129-7
eBook Packages: Springer Book Archive