Automatic facial feature detection is typically solved by usingmanually segmented imagestotrain a feature detector. In thispaper, we investigate whether it is possible toimprove the detection performance of such afeature detector by using additional unsegmented images. We propose a new adaptive automatic facial feature segmentation algorithm which aims to do this. The experimental results using this algorithmdemonstrate thatit is possible to improve the detection performance obtained from a small segmented training set by usinga larger number of additionalunsegmented images.
Hasan Demirel, Thomas J. Clarke, Peter Y. K. Cheun