Journal of Computer Science and Technology

, Volume 19, Issue 5, pp 684–690 | Cite as

Unified model in identity subspace for face recognition



Human faces have two important characteristics: (1) They are similar objects and the specific, variations of each face are similar to each other; (2) They are nearly bilateral symmetric. Exploiting the two important properties, we build a unified model in identity subspace (UMIS) as a novel technique for face recognition from only one example image per person. An identity subspace spanned by bilateral symmetric bases, which compactly encodes identity information. is presented. The unified model, trained on an obtained training set with multiple samples per class from a known people groupA, can be generalized well to facial images of unknown individuals, and can be used to recognize facial images from an unknown people groupB with only one sample per subject. Extensive experimental results on two public databases (the Yale database and the Bern database) and our own database (the ICT-JDL database) demonstrate that the UMIS approach is significantly effective and robust for face recognition.


pattern recognition face recognition identity subspace unified model 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Turk M, Pentland A. Eigenfaces for recognition.Journal of Cognitive Neuroscience, 1991, 3(1): 71–86.CrossRefGoogle Scholar
  2. [2]
    Edwards G, Taylor C J, Cootes T. Learning to identify and track faces in image sequences. InProc. Sixth IEEE Int. Conf. Computer Vision, Bombay, India, 1998, pp. 317–322.Google Scholar
  3. [3]
    Costen N, Cootes T, Edwards G, Taylor C J. Automatic extraction of the face identity-subspace.Image and Vision Computing, 2002, 20: 319–329.Google Scholar
  4. [4] Scholar
  5. [5] Scholar
  6. [6]
    Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection.IEEE Trans. Pattern Analysis and Machine Intelligence, 1997, 19(7): 711–720.CrossRefGoogle Scholar
  7. [7]
    Swets D L, Weng J. Using discriminant eigenfeatures for image retrieval.IEEE Trans. Pattern Analysis and Machine Intelligence, 1996, 18(8): 831–836.CrossRefGoogle Scholar
  8. [8]
    Etemad K, Chellappa R. Discriminant analysis for recognition of human face images.Journal of the Optiical Society of America A, 1997, 14: 1723–1733.Google Scholar
  9. [9]
    Moghaddam Baback, Jebara Tony, Pentland Alex. Bayesian face recognition.Pattern Recognition, 2000, 33(11): 1771–1782.CrossRefGoogle Scholar
  10. [10]
    Liu Chengjun, Wechsler Harry. Probabilistic reasoning models for face recognition.IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Santa Barbara, CA, USA, 1998, pp. 827–832.Google Scholar
  11. [11]
    Duvdevani-Bar Set al. A similarity-based method for the generalization of face recognition over pose and expression. InProc. 3rd IEEE Int. Conf. Automatic Face and Gesture Recognition (FG'1998), Nara, Japan, 1998, pp. 118–123.Google Scholar
  12. [12]
    Kirby M, Sirovich L. Application of the Karhunen-Loeve procedure for the characterization of human faces.IEEE Trans. Pattern Analysis and Machine Intelligence, 1990, 12(1): 103–108.CrossRefGoogle Scholar
  13. [13]
    Penev P S, Sirovich L. The global dimensionality of face space. InProc. 4th IEEE Int. Conf. Automatic Face and Gesture Recognition (FG'2000), Grenoble, France, 2000, pp. 264–270.Google Scholar

Copyright information

© Science Press, Beijing China and Allerton Press Inc., Beijing China and Allerton Press Inc. 2004

Authors and Affiliations

  • Pin Liao
    • 1
  • Li Shen
    • 1
  • Yi-Qiang Chen
    • 1
  • Shu-Chang Liu
    • 2
  1. 1.Institute of Computing TechnologyThe Chinese Academy of SciencesBeijingP.R. China
  2. 2.Multimedia Information Technology Teaching Center, School of InformationBeijing University of Posts and TelecommunicationsBeijingP.R. China

Personalised recommendations