This work aims to classify the changes in head pose of a user sitting in front of a screen by using the estimated head rotation. Considered classes include ∓15, ∓30, ve ∓ 45 degree pan, tilt and combinations of these poses. SIFT flow algorithm is used for motion estimation. Two dimensional feature vectors are extracted by calculating the magnitude and the angle of the flow vectors. Classification has been performed by Support Vector Machine and Naive Bayesian classifiers. Test results reported on Pointing'04 database demonstrate that SIFT flow vectors enable us classifying head rotation with high accuracy, when the desired resolution is not in the order of degrees.