Automatic model based face feature detection system

Abstract

Face recognition has become an important area of research in recent years, especially for security purposes. In this paper, we present a method to detect important features of the face for alignment purposes in a face recognition system. These features, which are the eye corners, mouth corners and nose tip, are detected using a combination of 2D and 3D images. The 2D image is used to locale the area of the eyes, mouth and nose while the 3D image is used to locate the exact point of the corners and tip. This method successfully detected the feature points at different face angles and is a fully automatic system.