A Combination of Spatiotemporal ICA and Euclidean Features for Face Recognition

  • Jiajin Lei
  • Tim Lay
  • Chris Weiland
  • Chao Lu
Part of the IFIP International Federation for Information Processing book series (IFIPAICT, volume 217)


ICA decomposes a set of features into a basis whose components are statistically independent. It minimizes the statistical dependence between basis functions and searches for a linear transformation to express a set of features as a linear combination of statistically independent basis functions. Though ICA has found its application in face recognition, mostly spatial ICA was employed. Recently, we studied a joint spatial and temporal ICA method, and compared the performance of different ICA approaches by using our special face database collected by AcSys FRS Discovery system. In our study, we have found that spatiotemporal ICA apparently outperforms spatial ICA, and it can be much more robust with better performance than spatial ICA. These findings justify the promise of spatiotemporal ICA for face recognition. In this paper we report our progress and explore the possible combination of the Euclidean distance features and the ICA features to maximize the success rate of face recognition.


Machine vision Face recognition Spatiotemporal ICA 


  1. [1]
    W. Zhao, R. Chellappa, P. J. Phillips, and A. Rosenfeld, Face Recognition: A Literature Survey, ACM Computing Surveys, Vol.35, No. 4, 399–458, December, 2003.CrossRefGoogle Scholar
  2. [2]
    Jongsun Kim, Jongmoo Choi, Juneho Yi, and Matthew Turk, Effective Representation Using ICA for Face Recognition Robust to Local Distortion and partial Occlusion, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27, No. 12, 2005, 1977–1981.CrossRefGoogle Scholar
  3. [3]
    M. Turk, A. Pentland, Eigenfaces for Recognition, Journal of Cognitive Neuroscience 3(1), 71–86, 1991.CrossRefGoogle Scholar
  4. [4]
    R. Brunell and T. Poggio, Face Recognition: Features vs. Templates, IEEE Trans. Pattern Analysis and Machine Intelligence, 15(10): 1042–1053, 1993.CrossRefGoogle Scholar
  5. [5]
    Chengjun Liu and Harry Wechsler, Comparative Assessment of Independent Component Analysis (ICA) for Face Recognition, In: the 2nd International Conference on Audio-and Video-Based Biometric Person Authentication, AVBPA’99, Washington D.C. USA, March 22–24,1999.Google Scholar
  6. [6]
    M. Stewart Bartlett, J. R. Movellan, and T. J. Sejnowski, Face Recognition by Independent Component Analysis, IEEE Transactions on Neural Network, Vol.13, Nov., 1450–1464, 2002.CrossRefGoogle Scholar
  7. [7]
    Victor C. Chen, “Spatial and Temporal Independent Component Analysis of Micro-Doppler Features” In: 2005 IEEE International Radar Conference Record, 348–353, 9–12 May 2005, Arlington, VA, USA.Google Scholar
  8. [8]
    Umbaugh, Scott E. Computer Imaging: Digital Image Analysis and Processing. New York, Taylor & Francis,2005.zbMATHGoogle Scholar
  9. [9]
    Bruce A. Draper, Kyungim Baek, Marian S. Bartlett, and J. Ross Beveridge, Recognizing Faces with PCA and ICA, http:/ Google Scholar
  10. [10]
    Andreas Jung, An Introduction to a New Data Analysis Tool: Independent Component Analysis, http://andreas.vvelcomes-you.CQm/research/paper/Jung_Intro_ICA_2002,pdf.Google Scholar
  11. [11]
    FastICA MATLAB package: Google Scholar
  12. [12]
    James V. Stone, Independent Component Analysis: A Tutorial Introduction, Bradford Book, 2004.Google Scholar
  13. [13]
    R.P.W. Duin, P. Juszczak, P. Paclik, E. Pekalska, D. de Ridder, D.M.J. Tax, Prtools, Scholar

Copyright information

© International Federation for Information Processing 2006

Authors and Affiliations

  • Jiajin Lei
    • 1
  • Tim Lay
    • 1
  • Chris Weiland
    • 1
  • Chao Lu
    • 1
  1. 1.Department of Computer and Information SciencesTowson UniversityTowsonUSA

Personalised recommendations