Tracking Facial Features Using Mixture of Point Distribution Models

  • Atul Kanaujia
  • Yuchi Huang
  • Dimitris Metaxas
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4338)


We present a generic framework to track shapes across large variations by learning non-linear shape manifold as overlapping, piecewise linear subspaces. We use landmark based shape analysis to train a Gaussian mixture model over the aligned shapes and learn a Point Distribution Model(PDM) for each of the mixture components. The target shape is searched by first maximizing the mixture probability density for the local feature intensity profiles along the normal followed by constraining the global shape using the most probable PDM cluster. The feature shapes are robustly tracked across multiple frames by dynamically switching between the PDMs. Our contribution is to apply ASM to the task of tracking shapes involving wide aspect changes and generic movements. This is achieved by incorporating shape priors that are learned over non-linear shape space and using them to learn the plausible shape space. We demonstrate the results on tracking facial features and provide several empirical results to validate our approach. Our framework runs close to real time at 25 frames per second and can be extended to predict pose angles using Mixture of Experts.


Gaussian Mixture Model Facial Feature Mahalanobis Distance Head Rotation Shape Space 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Cootes, T.: An Introduction to Active Shape Models. Oxford University Press, Oxford (2000)Google Scholar
  2. 2.
    Sozou, P., Cootes, T., Taylor, C., Dimauro, E.: Non-linear point distribution modelling using a multi-layer perceptron. BMVC (1995)Google Scholar
  3. 3.
    Cootes, T., Taylor, C.: A mixture model for representing shape variation. BMVC (1997)Google Scholar
  4. 4.
    Murase, H., Nayar, S.: Learning and recognition of 3D Objects from appearance. IJCV (1995)Google Scholar
  5. 5.
    Gong, S., Ong, E.J., McKenna, S.: Learning to associate faces across views in vector space of similarities to prototypes. BMVC (1998)Google Scholar
  6. 6.
    Romdhani, S., Gong, S., Psarrou, A.: A Multi-View Nonlinear Active Shape Model Using Kernel PCA. BMVC (1999)Google Scholar
  7. 7.
    Zhou, Y., Zhang, W., Tang, X., Shum, H.: A Bayesian Mixture Model for Multi-view Face Alignment. CVPR (2005)Google Scholar
  8. 8.
    Cootes, T., Wheeler, G., Walker, K., Taylor, C.: View-Based Active Appearance Models. BMVC (2001)Google Scholar
  9. 9.
    Edwards, G.J., Taylor, C.J., Cootes, T.F.: Learning to Identify and Track Faces in Image Sequences. BMVC (1997)Google Scholar
  10. 10.
    Heap, T., Hogg, D.: Improving specificity in pdms using a hierarchical approach. BMVC (1997)Google Scholar
  11. 11.
    Bregler, C., Omohundro, S.: Surface Learning with Applications to Lipreading. NIPS (1994)Google Scholar
  12. 12.
    Goodall, C.: Procrustes methods in the statistical analysis of shape. Journal of the Royal Statistical Society (1991)Google Scholar
  13. 13.
    Tomasi, C., Kanade, T.: Detection and Tracking of Point Features. Technical Report CMU-CS-91-132 (1997)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Atul Kanaujia
    • 1
  • Yuchi Huang
    • 1
  • Dimitris Metaxas
    • 1
  1. 1.Department of Computer ScienceRutgers University 

Personalised recommendations