Latent Identity Variables: Biometric Matching Without Explicit Identity Estimation

  • Simon J. D. Prince
  • Jania Aghajanian
  • Umar Mohammed
  • Maneesh Sahani
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4642)


We present a new approach to biometrics that makes probabilistic inferences about matching without ever estimating an identity “template”. The biometric data is considered to have been created by a noisy generative process. This process consists of (i) a deterministic component, which depends entirely on an underlying representation of identity and (ii) a stochastic component which accounts for the fact that two biometric samples from the same person are not identical. In recognition, we make inferences about whether the underlying identity representation is the same without ever estimating it. Instead we treat identity as fundamentally uncertain and consider all possible values in our decision. We demonstrate these ideas with toy examples from face recognition. We compare our approach to the class-conditional viewpoint.


Biometrics Face Recognition Bayesian Methods 


  1. 1.
    Belhumeur, P.N., Hespanha, J., Kriegman, D.J.: Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection. PAMI 19, 711–720 (1997)Google Scholar
  2. 2.
    Blanz, V., Romdhani, S., Vetter, T.: Face identification across different poses and illumination with a 3D morphable model. In: ICFGR, pp. 202–207 (2002)Google Scholar
  3. 3.
    Bolle, R.M., Connell, J.H., Pankathi, S., Ratha, N.K., Senior, A.W.: Guide to Biometrics. Springer, New York (2004)Google Scholar
  4. 4.
    Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood for incomplete data via the EM algorithm. Proc. Roy. Stat. Soc. B 39, 1–38 (1977)zbMATHMathSciNetGoogle Scholar
  5. 5.
    Fei-Fei, L., Fergus, R., Perona, P.: A Bayesian approach to unsupervised one-shot learning of object categories. In: CVPR, pp. 1134–1141 (2003)Google Scholar
  6. 6.
    Jordan, M.I., Ghahramani, Z., Jaakola, T.S., Saul, L.K.: An introduction to variational methods for graphical models. In: Jordan, M.I. (ed.) Learning in Graphical Models (1998)Google Scholar
  7. 7.
    MacKay, D.: Information theory, inference, and learning algorithms. Cam. U. Press, Cambridge (2003)zbMATHGoogle Scholar
  8. 8.
    Moghaddam, B., Jebra, T., Pentland, A.: Bayesian face recognition. Pattern Recognition 3, 1771–1782 (2000)CrossRefGoogle Scholar
  9. 9.
    Prince, S.J.D., Elder, J.H.: Tied factor analysis for face recognition across large pose changes. In: BMVC, vol. 3, pp. 889–898 (2006)Google Scholar
  10. 10.
    Turk, M., Pentland, A.P.: Face recognition Using eigenfaces. In: CVPR, pp. 586–591 (1991)Google Scholar
  11. 11.
    Zhou, S.K., Chellappa, R.: Probabilistic Identity characterization for face recognition. In: CVPR, vol. 2, pp. 805–812 (2004)Google Scholar
  12. 12.

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Simon J. D. Prince
    • 1
  • Jania Aghajanian
    • 1
  • Umar Mohammed
    • 1
  • Maneesh Sahani
    • 2
  1. 1.Computer Science 
  2. 2.Gatsby Unit, University College LondonUK

Personalised recommendations