An automatic depth map generation for 2D-to-3D conversion

In this paper, we propose an automatic algorithm which generates depth map of 2D image in good quality. The depth map is generated by fusing global depth gradient and local depth refinement. Image classification technique and edge-based depth cue are used to obtain the global depth gradient. In addition, image segmentation method and warm/cool color theory are adopted to generate local depth refinement. The proposed algorithm has advantages of good quality and low complexity. After the depth map is generated, the original 2D image or video can be converted to 3D for showing on stereoscopic display devices.