手振り認識システムのための幅情報を用いたカメラ位置・姿勢推定

抄録

We propose an automatic camera calibration method to determine the position and orientation parameters of a newly installed camera in our hand gesture tracking system. In a multiple camera-based system, automatic camera calibration algorithms are required due to the increasing cost of camera calibration. In our method, both 3D position tracking results and 2D positions and sizes on camera image planes are used for the calibration. By using 2D width information for camera calibration, the calibration results stabilize even if the input data includes large observation errors. Observations by the newly installed camera are taken into account while tracking to increase the reliability of the calibration result. Using this method, the cost of installing new cameras in a hand tracking system can be effectively reduced and we can easily add new camera (s) to the hand gesture tracking system. Moreover, the hand tracking accuracy can be improved and the tracking area can be widened by camera addition. Experimental results show the effectiveness of the proposed method.