Authors:
Jeong-ho Kim
1
;
Dae-woo Lee
1
;
Se-jong Heo
2
;
Chan-gook Park
2
;
Kwang-yul Baek
3
and
Hyo-choong Bang
3
Affiliations:
1
Pusan National University, Korea, Republic of
;
2
Seoul National University, Korea, Republic of
;
3
KAIST, Korea, Republic of
Keyword(s):
Head-tracker, Eye-tracker, Integration, Sensor fusion, Dichroic filter.
Related
Ontology
Subjects/Areas/Topics:
Human-Robots Interfaces
;
Image Processing
;
Informatics in Control, Automation and Robotics
;
Robotics and Automation
;
Sensors Fusion
;
Signal Processing, Sensors, Systems Modeling and Control
;
Vision, Recognition and Reconstruction
Abstract:
This paper describes the development of an integrated head/eye tracker system. To obtain the position and attitude of a head, the sensor-fusion head tracker is used. The head tracker combines the result of the vision-based tracking and the IMU to increase the tracking accuracy. Five sets of IR LEDs are installed on the surface of a helmet, and the IMU is installed inside the helmet; each set of LEDs comprises three LEDs positioned at the vertices of a triangle. IR LEDs are used on the eye-tracker system since they are more suitable than visible LEDs for cognizing the pupil. For the precise tracking of the pupil, three methods—intensity-based detection method, shape-based detection method, and sequential mean-shift method—are used. The gaze vector is calculated by using the obtained position of the pupil, focal length, and gaze-point equation. Finally, we verify whether this integrated system can be used in practical military equipment.