Offline Camera Movement Tracking from Video Sequences

In this paper, we propose a method to track the movement of camera from the video sequences. This method is useful for video analysis and can be applied as pre-processing step in some application such as video stabilizer and marker-less augmented reality. First, we extract the features in each frame using corner point detection. The features in current frame are then compared with the features in the adjacent frames to calculate the optical flow which represents the relative movement of the camera. The optical flow is then analyzed to obtain camera movement parameter. The final step is camera movement estimation and correction to increase the accuracy. The method performance is verified by generating a 3D map of camera movement and embedding 3D object to the video. The demonstrated examples in this paper show that this method has a high accuracy and rarely produce any jitter.