Title :
A face recognition method robust to pose variation
Author :
Ying, Wang ; Lifang, Wu ; Ling, Tu ; Xue, Wu
Author_Institution :
Sch. of Electron. Inf. & control Eng., Beijing Univ. of Technol., Beijing
Abstract :
In this paper, a face recognition method robust to pose variation is proposed. First, Cascade-MR-ASM is utilized to locate feature points in face image, then the location results are mapped into the public parameter space and the shape parameters are obtained. Then the frontal ASM model is obtained by setting zero to parameters related to pose. Then frontal face image is obtained by texture mapping. Finally, face recognition based on eigenface is finished using the K-Nearest Neighbor classifier. The experimental results in CMU_PIE show that the algorithm is robustness to variation of pose.
Keywords :
eigenvalues and eigenfunctions; face recognition; feature extraction; image classification; image texture; pose estimation; shape recognition; K-nearest neighbor classifier; cascade multiresolution active shape model; eigenface; face image recognition method; image texture mapping; pose variation; public parameter space; Active appearance model; Active shape model; Clustering algorithms; Control engineering; Face recognition; Lighting; Robust control; Robustness; Shape control; Testing;
Conference_Titel :
Signal Processing, 2008. ICSP 2008. 9th International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-2178-7
Electronic_ISBN :
978-1-4244-2179-4
DOI :
10.1109/ICOSP.2008.4697441
