When the user initially clicks the eye feature, a box is drawn around the square and the subimage within this square is cropped out of the image frame. The cropped image is used as a template to find the position of the feature in subsequent frames. The system determines the position of the feature using Sum of Square Differences (SQD) method. To reduce extensive computation, the system tracks the feature in a "search window", a small area around the position of the feature in previous frame.