A Robust 3D Face Recognition Algorithm Using Passive Stereo Vision

Akihiro HAYASAKA  Koichi ITO  Takafumi AOKI  Hiroshi NAKAJIMA  Koji KOBAYASHI  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E92-A   No.4   pp.1047-1055
Publication Date: 2009/04/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E92.A.1047
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Advanced Technologies Emerging Mainly from the 21st Workshop on Circuits and Systems in Karuizawa)
Category: 
Keyword: 
biometrics,  face recognition,  3D face,  facial expression,  phase-only correlation,  

Full Text: PDF(3.1MB)>>
Buy this Article




Summary: 
The recognition performance of the conventional 3D face recognition algorithm using ICP (Iterative Closest Point) is degraded for the 3D face data with expression changes. Addressing this problem, we consider the use of the expression-invariant local regions of a face. We find the expression-invariant regions through the distance analysis between 3D face data with the neutral expression and smile, and propose a robust 3D face recognition algorithm using passive stereo vision. We demonstrate efficient recognition performance of the proposed algorithm compared with the conventional ICP-based algorithm through the experiment using a stereo face image database which includes the face images with expression changes.