Hybrid visual tracking and model-based alignment using 6-DOF sensor

In this paper, we propose a hybrid visual tracking and model-based alignment using a six degree of freedom (6-DOF) sensor for augmented reality (AR). We use a head mounted display (HMD) with a 6-DOF sensor as user's vision, and to acquire the corresponding data for being able to estimate the user's head pose and movement. We combine use of the 6-DOF sensor and image processing. By using amount of the acceleration change and angular data obtained by the sensor, the target position and pose are estimated. Specifically, we track the alignment target using particle filter whose particles are sampled on the basis of the acquired change data from the sensor. Moreover, the target is tracked with a simplified plane model. In order to show the effectiveness of the proposed system, we conduct the experiment of tracking and model-based alignment when we applied the system to real space.