Automatic facial feature points detection

Abstract

This paper presents an algorithm which detects automatically the feature points in a face image. This is a fundamental task in many applications, in particular in an automatic face recognition system. Starting from a frontal face image with a plain background we have effected an image segmentation to detect the different facial components (eyebrow, eyes, nose, mouth and chin). After this we have searched for the feature points of each face component. The algorithm has been tested on 320 face images taken from the Stirling University Face Database [10]. The points extracted in this way have been used in a face recognition algorithm based on the Hough transform. © 2008 Springer-Verlag Berlin Heidelberg.