Conversion of 2D Images to 3D

In recent years, 3D image display has become a trend in the visual processing field. 3D displays provide better visual quality than conventional 2D displays. The depth information needed for the 3D display is not available in the 2D content. This paper presents an algorithm for the automatic conversion of 2D images to 3D based on the depth information. This algorithm uses a single 2D image for the conversion process. The image is sectioned into number of blocks based on pixel intensity and edge detection. Then the depth values are assigned as per the hypothesis depth value. Cross bilateral filter is then used to enhance the visual comfort. Then the filtered image is processed using a Depth Image Based Rendering (DIBR) method to generate the 3D image. The key step in the process is the generation of depth map since the visual quality of the reconstructed image depends on the dense depth information.