A new eye location method based on Ring Gabor Filter
Yanfang Zhang, Nongliang Sun, Yang Gao, Maoyong Cao · 2008
The eyes are the most important facial landmarks on human face. It is significant to extract the feature of eyes efficiently. In order to improve the real-time and veracity performance of the eye location method, this paper presents a coarse-to-fine eye location method based on ring Gabor filter (RGF). First an elliptical Gabor filter is devised to convolute with the face image to highlight the eye region, which can be exploited to segment the two pupil regions efficiently. After getting the region of the pupil, a circular Gabor filter is used to locate the centre of the eyeball. Extensive experiments on the face base show that the method can locate the pupils accurately. Comparing with other algorithms, results prove that this method is robust to the variations of face poses, expressions, accessories and illuminations.