Many researches about eye tracking have been done in field of human computer interface, virtual reality, and so on. However, most of eye tracking devices are priced high because of its complicated hardware. Hence, in this paper, we propose a simple real-time eye tracking system using web camera. In our system, we use a zoom lens for web camera, which cover the low resolution of eye region by wide-view capturing. In addition, an infrared light illuminator is required because the gaze position is calculated by using the positional relationship between the pupil and specular reflection caused by the illuminator. Our system measures the gaze position using just one eye, and does not require to wear any devices onto user’s head. The average error of gaze point estimation was about 0.9 degree at the Z distance of 70cm on the 23inch size monitor. Experimental results showed that the proposed system can be used enough for navigation or pointing interaction.
[1]
Kang Ryoung Park,et al.
A Study on Eye Gaze Estimation Method Based on Cornea Model of Human Eye
,
2007,
MIRAGE.
[2]
Kang Ryoung Park,et al.
Binocular gaze detection method using a fuzzy algorithm based on quality measurements
,
2014
.
[3]
Y. Matsumotot,et al.
Development of intelligent wheelchair system with face and gaze based interface
,
2001,
Proceedings 10th IEEE International Workshop on Robot and Human Interactive Communication. ROMAN 2001 (Cat. No.01TH8591).
[4]
Matthew Banta,et al.
What do exploratory searchers look at in a faceted search interface?
,
2009,
JCDL '09.
[5]
Zhiwei Zhu,et al.
Eye and gaze tracking for interactive graphic display
,
2002,
SMARTGRAPH '02.