Skip to Main Content
A multimodal part face recognition method based on principal component analysis (MMP-PCA) is presented. Five kinds of parts are dealt with by PCA to obtain eigenface, eigeneyebrow, eigeneye, eigennose and eigenmouth. Thirty-one kinds of different face recognition models are created by using combinations of different parts. The recognition rates of 17 models among them are given.
Date of Publication: 5 Dec 2002