Depth map extracting based on geometric perspective: An applicable 2D to 3D conversion technology

To alleviate the dilemma of 3D resources shortage, a method is claimed in this paper. This method analyzes the direction and intersection of the major segments in the image scene. Based on the geometrical elements existing in the general scene, we get vanishing line and vanishing point. Taken the vanishing point's coordinate as a centre position, each pixel point has its own distance. According to this distance, the corresponding gray value of each pixel point is calculated. For each pixel point, this is the relative depth. At last, we can generate a whole depth map for the original image. The generated depth map and the original 2D image can be converted to the 3D composite display, by using the virtual view rendering technique based on the depth map.