A method for estimating the depth information of a general monocular image sequence and then creating a 3D stereo video is proposed. Distinguishing between foreground and background is possible without additional information, and then foreground pixels ar
