Title :
Lower bounds estimation to KL transform in face representation and recognition
Author :
Zhang, Bai-bo ; Zhang, Chang-Shui
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing, China
Abstract :
The Karhunen-Loeve (KL) transform is a very important method in data compression and pattern recognition, which has been successfully used in face recognition. Intuitively, the more training samples are used, the better performance would be, but the computation cost increases in proportion to cubic speed. The numbers of training samples and eigenvalues required by the KL method in face representation and recognition are considered. In this paper, we attempt to find the balance points by experiments and analyses.
Keywords :
Karhunen-Loeve transforms; eigenvalues and eigenfunctions; face recognition; image representation; learning (artificial intelligence); KL transform; Karhunen-Loeve transform; computation cost; eigenvectors; face recognition; face representation; lower bounds estimation; training samples; Automation; Face recognition; Image databases; Intelligent systems; Karhunen-Loeve transforms; Magnetic heads; Mouth; Pattern recognition; Spatial databases; State estimation;
Conference_Titel :
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7508-4
DOI :
10.1109/ICMLC.2002.1167417
