Gaze tracking based on run length coding

Eye gaze determines the view point, focus of attention and interest of an individual. Gaze estimation systems compute the direction of eye gaze. These systems play an important role in human-computer interaction. The gaze pointing systems is a replacement for existing input devices in gaze communication applications. The gaze tracking methods are based on interpolation, 2D geometric transform mapping, edge change ratio in successive frames and singular value decomposition. This paper discusses a robust, less computational gaze tracking method using run length coding. 94.5% recognition rate of gaze direction is achieved using run length coding.