Towards practical inside-out head tracking for mobile seating bucks
Markus Miezal, Gabriele Bleser, Didier Stricker, Johannes Tümler
ISMAR 2012 Workshop on Tracking Methods and Applications Workshop on Tracking Methods and Applications (ISMAR-12), November 5-8, Atlanta, Georgia, USA

Abstract:

This paper proposes a conceptual solution and a preliminary study for a head tracking system designed to fulfill the strong require- ments of a mobile seating buck application, such as easy setup, min- imal latency, and wide effective field of view (FOV). The proposed approach uses a head-mounted camera-IMU (inertial measurement unit) system with an ultra-wide FOV fisheye lens for tracking the head with respect to one single circular marker visualized on the navigation display. The visual-inertial sensor fusion is performed with help of an extended Kalman filter. An image processing ap- proach and several fusion models are developed and their perfor- mances are evaluated and compared in relation to the specific tar- get application. The paper shares the experiences and experimental results from this preliminary study, as well as, ideas for improve- ments and extensions for the next development cycle. Overall, the study shows that the proposed setup in combination with a tightly coupled sensor fusion approach provides a feasible solution for the application context.

Towards practical inside-out head tracking for mobile seating bucks
(Hrsg.)
ISMAR 2012 Workshop on Tracking Methods and Applications Workshop on Tracking Methods and Applications (ISMAR-12), November 5-8, Atlanta, Georgia, USA

Abstract:

This paper proposes a conceptual solution and a preliminary study for a head tracking system designed to fulfill the strong require- ments of a mobile seating buck application, such as easy setup, min- imal latency, and wide effective field of view (FOV). The proposed approach uses a head-mounted camera-IMU (inertial measurement unit) system with an ultra-wide FOV fisheye lens for tracking the head with respect to one single circular marker visualized on the navigation display. The visual-inertial sensor fusion is performed with help of an extended Kalman filter. An image processing ap- proach and several fusion models are developed and their perfor- mances are evaluated and compared in relation to the specific tar- get application. The paper shares the experiences and experimental results from this preliminary study, as well as, ideas for improve- ments and extensions for the next development cycle. Overall, the study shows that the proposed setup in combination with a tightly coupled sensor fusion approach provides a feasible solution for the application context.