Camera Pose Estimation in Soccer Scenes Based on Vanishing Points and its Application in Virtual Advertisement Insertion

Camera parameters estimation is an important issue in machine vision. This paper proposes a new method to find translation and rotation matrix of camera in sport scene on the basis of vanishing points for the virtual advertisement insertion. Vanishing point (VP) of parallel lines is the image of the point at infinity, which corresponds to the projection of the intersection of parallel lines at infinity. According to projective geometry constraint, camera rotation of the projection matrix is computed directly by two vanishing points, then pan and tilt of camera extracted from this matrix. Computer simulation and real data experiments are carried out to validate our method. Experimental results have shown that this method can meet the needs of the virtual advertisement insertion and be put into use in real application scenarios.

Keywords:

Camera calibration , pose estimation , Rotation matrix , Soccer , vanishing points , virtualadvertisement