A Comparison of Different Face Recognition Algorithms

Yi-Hsin Liu Mark Ng Chun-Wei Liu

National Taiwan University

Teaser

The patches fined by Facial Trait Algorithm.

Abstract

We propose the Ensemble Voting Algorithm for face recognition. Our method outperformed four canonical algorithms. The algorithms train on a data set with 1815 faces, register on a gallery set with 1027 faces, and test on four different probe sets. This paper will detail the parameter tuning process, and report the testing result via receiver operating characteristic (ROC) curve and verification accuracy.

Download


Download the pdf file Paper:
8-page PDF

References

  1. “Face Recognition with Local Binary Patterns.” [Link]
    Timo Ahonen, Abdenour Hadid, and Matti Pietikäinen.
    ECCV 2004
  2. “Eigenfaces vs. fisherfaces: Recognition using class specific linear projection.” [Link]
    Peter N. Belhumeur, and Joao Pedro Hespanha, D.J. Kriegman.
    PAMI 1997
  3. “LIBSVM: a library for support vector machines.” [Link]
    Chih-Chung Chang, and Chih-Jen Lin.
    2001
  4. “Facial trait code and its application to face recognition.” [Link]
    Ping-Han Lee, Gee-Sern Hsu, Tsuhan Chen, and Yi-Ping Hung.
    ISVC 2008
  5. “Face recognition using eigenfaces.” [Link]
    Matthew Turk, and A. P. Pentland.
    CVPR 1991