3D Face Recognition Using Modified PCA Methods

In this paper we present an approach for 3D face recognition based on extracting principal components of range images by utilizing modified PCA methods namely 2DPCA and bidirectional 2DPCA also known as (2D) 2 PCA.A preprocessing stage was implemented on the images to smooth them using median and Gaussian filtering. In the normalization stage we locate the nose tip to lay it at the center of images then crop each image to a standard size of 100*100. In the face recognition stage we extract the principal component of each image using both 2DPCA and (2D) 2 PCA. Finally, we use Euclidean distance to measure the minimum distance between a given test image to the training images in the database. We also compare the result of using both methods. The best result achieved by experiments on a public face database shows that 83.3 percent is the rate of face recognition for a random facial expression.

Authors:

• Omid Gervei
• Ahmad Ayatollahi
• Navid Gervei

Keywords:

• 3D face recognition
• 2DPCA
• (2D) 2 PCA
• Rangeimage

References:

[1] S.Z. Li, and A.K Jain, "Handbook of Face Recognition", ISBN: 0-387-
40595-X, 2005.
[2] P. J. Phillips, P.J. Flynn, T. Scruggs, K. W. Bowyer, J. Chang, K.
Hoffman, J. Marques, J. Min, and W. Worek, "Overview of the face
recognition grand challenge," in IEEE Conference on Computer Vision
and Pattern Recognition, 2005.
[3] Beumier, C., M. Acheroy, "Automatic 3D face authentication,"Image
and Vision Computing, vol.18, no.4, pp.315-321, 2000.
[4] Wang, Y., C. Chua, Y. Ho, "Facial feature detection and face
recognition from 2D and 3D images," Pattern Recognition Letters,
vol.23, pp.1191-1202, 2002.
[5] Tsalakanidou, F., D. Tzovaras, M. Strinzis,"Use of depth and color
eigenfaces for face recognition", Pattern Recognition Letter, vol.24,
pp.1427-1435, 2003.
[6] M. H. Mahoor and M. Abdel-Mottaleb, "3D face recognition based on
3D ridge lines in range data "in ICIP,2007
[7] Hesher, C., A. Srivastava, G. Erlebacher, "A novel technique for face
recognition using range imaging," in Proc. 7th Int. Symposium on Signal
Processing and Its Applications, pp.201-204, 2003.
[8] Moreno, A.B., ' A. S'anchez, J.F. V'elez, F.J. D'─▒az, "Face recognition
using 3D surface-extracted descriptors," in Proc.IMVIPC,2003.
[9] A.B.Moreno, A.Sanchez, J.F.Velez, F.J.Dkaz, Face recognition using 3d
local geometrical features: PCA vs. SVM, in: Fourth International
Symposium on Image and Signal Processing and Analysis (ISPA2005),
2005.
[10] Otsu N.," A threshold selection method from gray-level histogram",
IEEE Trans. Syst. Man Cybern. 9:62-66; 1979.
[11] J. Yang, D. Zhang, A.F. Frangi, and J.Y. Yang, "Two-Dimensional
PCA: A New Approach to Appearance-Based Face Representation and
Recognition," IEEE Transactions on Pattern Analysis and Machine
Intelligence, Vol. 26, pp. 131-137, 2004.
[12] D. Zhang, Z.H. Zhou, (2D)2PCA: two-directional two-dimensional PCA
for efficient face representation and recognition, Neurocomputing 69,
pp. 224-231, 2005.
[13] Young-Gil Kim, Young-Jun Song, Un-Dong Chang, Dong-Woo Kim,
Tae-Sung Yun, Jae-Hyeong, "Face recognition using a fusion method
based on bidirectional 2DPCA", Applied Mathematics and Computation
205, pp. 601-607, 2008.
[14] A. B. Moreno, A. Sanchez, GavabDB: a 3D Face Database, in: Proc. 2nd
COST275 Workshop on Biometrics on the Internet (Vigo (Spain),
2004), 75-80, URL http://gavab.escet.urjc.es.