Valentin D, Abdi H, Edelman B, O'Toole AJ
The University of Texas at Dallas
J Math Psychol. 1997 Dec;41(4):398-413. doi: 10.1006/jmps.1997.1186.
We present an overview of the major findings of the principal component analysis (pca) approach to facial analysis. In a neural network or connectionist framework, this approach is known as the linear autoassociator approach. Faces are represented as a weighted sum of macrofeatures (eigenvectors or eigenfaces) extracted from a cross-product matrix of face images. Using gender categorization as an illustration, we analyze the robustness of this type of facial representation. We show that eigenvectors representing general categorical information can be estimated using a very small set of faces and that the information they convey is generalizable to new faces of the same population and to a lesser extent to new faces of a different population. Copyright 1997 Academic Press. Copyright 1997 Academic Press
我们概述了主成分分析（PCA）方法在面部分析中的主要发现。在神经网络或联结主义框架中，这种方法被称为线性自联想器方法。面部被表示为从面部图像的叉积矩阵中提取的宏观特征（特征向量或特征脸）的加权和。以性别分类为例，我们分析了这种面部表示类型的稳健性。我们表明，代表一般分类信息的特征向量可以使用非常少量的面部进行估计，并且它们所传达的信息可以推广到同一人群的新面孔，在较小程度上也可以推广到不同人群的新面孔。版权所有1997年学术出版社。版权所有1997年学术出版社
