Efficient uncalibrated rectification method for stereo vision systems

Along with the rapid development of stereoscopic cinematography in the last few years, efficient 3D video diagnostic tools are increasingly in demand. The tools dedicated for real-time video analysis and assessment directly on the movie set are aimed at facilitation, acceleration and cost-cutting of movie production. It causes the growing interest in development of high-performance and reliable algorithms for stereoscopic video processing. One of the algorithms strictly related to 3D video processing is image rectification. It is aimed at transformation of the stereopair images in such a way that epipolar lines are horizontal and match up between views. The rectification is often a required preprocessing step for depth estimation algorithms because it simplifies searching for matching points in stereo images. The paper presents an overview and comparison of image rectification methods for the application in real-time diagnostic tools for stereoscopic cinematography. It also describes the implementation and evaluation of the rectification algorithm based on [1] that has been finally selected for the needs of diagnostic tool for stereo rig calibration and disparity map estimation developed by the authors. Described algorithm ensures very accurate image alignment and minimization of vertical disparity for the needs of disparity map calculation. Moreover, it is resistant to lens distortions in input image and does not introduce significant distortions in output image preserving the original viewpoint of the cameras.