Converting facial expressions using recognition-based analysis of image Sequences

  • Takahiro Otsuka
  • Jun Ohya
Poster Session III
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1352)


A method for converting one person's facial expression into another person's is proposed. The sequence of the feature vector for each expression is modeled by using HMM with the hidden states corresponding to the different muscle conditions (relaxed, contracting, and the end of contraction). The probability of each state is evaluated for each frame and the contraction rate of each muscle is obtained from the probability of each state using a matrix representing the characteristics of other people's expressions. The experiments showed the superior realism of the expression generated by our proposed method.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    J. Ohya, K. Ebihara, J. Kurumisawa, and R. Nakatsu, “Virtual Kabuki Theater: Towards the Realization of Human Metamorphosis Systems,” Proc. IEEE International Workshop on Robot and Human Communication, pp. 416–421, Nov. 1996.Google Scholar
  2. 2.
    D. Terzopoulos, and K. Waters, “Analysis and Synthesis of Facial Image Sequences Using Physical and Anatomical Models,” IEEE Rans. PAMI, vol. 15, no. 6, pp. 569–579, 1993.Google Scholar
  3. 3.
    I. A. Essa, “Analysis, Interpretation, and Synthesis of Facial Expressions,” M.I.T. Media Laboratory, Perceptual Computing Group Report no. 303, 1994.Google Scholar
  4. 4.
    J. Ohya, Y. Kitamura, H. Ishii, H. Takemura, F. Kishino, and N. Terashima, “Virtual Space Teleconferencing: Real-time Reproduction of 3-D Human Image,” Journal of Visual Communication and Image Representation (VCIR), vol. 6, no. 1, pp. 1–25, March 1995.Google Scholar
  5. 5.
    T. Otsuka, and J. Ohya, “Recognition of Racial Expressions Using HMM with Continuous Output Probabilities,” Proc. IEEE International Workshop on Robot and Human Communication, pp. 323–328, Nov. 1996.Google Scholar
  6. 6.
    P. Ekman, and W. V. Friesen, “The Facial Action Coding System,” Consulting Psychologists Press, Inc., 1978.Google Scholar
  7. 7.
    S. M. Platt, and N. I. Badler, “Animating Facial Expressions,” Computer Graphics, vol. 15, no. 3, pp. 245–252, 1981.Google Scholar
  8. 8.
    K. Waters, “A Muscle Model for Animating Three-Dimensional Facial Expression,” Comput. Graphics, vol. 22, no. 4, pp. 17–24, 1987.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Takahiro Otsuka
    • 1
  • Jun Ohya
    • 1
  1. 1.ATR Media Integration & Communications Research LaboratoriesKyotoJapan

Personalised recommendations