This paper presents the method to determine the order of feature (attention) points for face detection. The order of feature points is determined in terms of the classification ability to face and non-face images (cross validation). The matching is performed in the order of the selected feature points. It is confirmed that high recognition rate is obtained by using only a small number of feature points. By ordering of the feature points, we can reject the input image as non-face by systematically checking the feature points. This can speed up the face detection.