3D Image format identification by image difference

Abstract

Many 3D formats exist and will co-exist for a long time since there is no 3D standard that defines a generally accepted 3D format. The support for multiple 3D formats will be important for bringing 3D into home. In this paper, we propose a fast and effective method to detect whether an image is a 2D image or a 3D image encoded with a pair of stereo images, and to further identify the exact 3D format in the latter case. The proposed method computes an edge map resulted from image differences between the left and right view images; uses the statistics from the distribution of edge widths combined with structure similarity analysis to detect the existence of a 3D format and to identify the format. Experiments show the effectiveness of this method.