Improving the estimation of head pose orientation: By using eyeglasses as a key feature

The paper refines and improves an unique approach for estimating the head pose orientation of subjects wearing eyeglasses by using a combination of line detection and classification approaches by extending from previously, limited discrete face pose, into more accurate face poses based the relative pitch and yaw face orientation. The estimation of the face pose orientation could be used in non-verbal communication, in the area of human-computer interface, and in related areas.