Title :
Application of the Karhunen-Loeve procedure for the characterization of human faces
Author :
Kirby, M. ; Sirovich, L.
Author_Institution :
Div. of Appl. Math., Brown Univ., Providence, RI, USA
fDate :
1/1/1990 12:00:00 AM
Abstract :
The use of natural symmetries (mirror images) in a well-defined family of patterns (human faces) is discussed within the framework of the Karhunen-Loeve expansion. This results in an extension of the data and imposes even and odd symmetry on the eigenfunctions of the covariance matrix, without increasing the complexity of the calculation. The resulting approximation of faces projected from outside of the data set onto this optimal basis is improved on average
Keywords :
eigenvalues and eigenfunctions; matrix algebra; pattern recognition; Karhunen-Loeve expansion; Karhunen-Loeve procedure; covariance matrix; eigenfunctions; human faces; mirror images; natural symmetries; pattern recognition; Covariance matrix; Degradation; Eigenvalues and eigenfunctions; Face recognition; Humans; Linear regression; Mathematics; Mirrors; Neural networks; Speech;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
