Automatic and Robust Detection of Facial Features in Frontal Face Images

Abstract

Automatic detection of facial features in an image is important stage for various facial image interpretation work, such as face recognition, facial expression recognition, 3Dface modeling and facial features tracking. Detection of facial features like eye, pupil, mouth, nose, nostrils, lip corners, eye corners etc., with different facial expression and illumination is a challenging task. In this paper, we presented different methods for fully automatic detection of facial features. Viola-Jones' object detector along with haar-like cascaded features are used to detect face, eyes and nose. Novel techniques using the basic concepts of facial geometry, are proposed to locate the mouth position, nose position and eyes position. The estimation of detection region for features like eye, nose and mouth enhanced the detection accuracy significantly. An algorithm, using the H-plane of the HSV color space is proposed for detecting eye pupil from the eye detected region. FEI database of frontal face images is mainly used to test the algorithm. Proposed algorithm is tested over 100 frontal face images with two different facial expression (neutral face and smiling face). The results obtained are found to be 100% accurate for lip, lip corners, nose and nostrils detection. The eye corners, and eye pupil detection is giving approximately 95% accurate results.