Fully automatic and quickly facial feature point detection based on LK algorithm

To identify people\´s facial expressions, locating facial feature points in images of faces is an important stage. In this paper we present a method for fully automatic and quick detection of facial feature points in image sequences or videos using Gabor feature based boosted classifiers based on LK algorithm. There are many ways detecting face in complex background, we assume that the face has been identified. The detected face region is divided into several regions of interest, each of which contains one feature point or more. We selected several "best feature points" in every region of interest used LK algorithm. The proposed facial feature point detection method uses individual feature patch templates to detect points in these "best feature points". These feature models are GentleBoost templates built from both gray level intensities and Gabor wavelet features. We tested our method with several videos, and the method has achieved a recognition rate of 95%.