Publications

Abstract : The PCCR (Pupil Center Corneal Reflection) method became dominant for finding human's diverse eye gaze directions through the research on the eye tracking technology that has been done for a very long period of time. The initial studies on the eye tracking technology were related to the general human interface for operating equipment and devices, then it has been promoted to the field of various purposes such as a market research in a recent study analyzing customer's behaviors. In particular, a real time eye gaze tracking system is most important for many HCI applications including stereoscopic synthesis, intend extraction, behavior analysis and etc. In order to make an eye gaze tracking system to be real time, the system must have an efficient pupil detection algorithm and ambience-independent image processing as well as reduced complexity, small size and number of circuit components. This paper proposes a method for getting clean images compared to the previous systems to reduce image processing overhead. Because it also helps reducing the number of image frames to be dropped during the image processing, the proposed method can provide a sufficient performance even on a low cost hardware system by reducing the transmission traffic.