Abstract
In this paper, we propose a new Augmented Reality method that can render CG models on the live video under marker-free condition by memorizing the marker information. In order to realize this system, it is necessary to obtain the camera posture information as well as the marker position in the world coordinate so that the system uses an inertial sensor to obtain the camera posture. With this method, the proposed system can render CG models, which size does not depend on the marker, even after the marker disappears.
