3D-TV content generation: 2D-to-3D conversion

The next major advancement in television is expected to be stereoscopic three-dimensional television (3D-TV). A successful roll-out of 3D-TV will require a backward-compatible transmission and distribution system, inexpensive 3D displays that are equal or superior to high-definition television (HDTV), and an adequate supply of high-quality 3D program material. With respect to the last factor, we reckon that the conversion of 2D material to stereoscopic 3D could play an important role. In this paper we provide (a) an overview of the fundamental principle underlying 2D-to-3D conversion techniques, (b) a cursory look at a number of approaches for depth extraction using a single image, and (c) a highlight of the potential use of surrogate depth maps in depth image based rendering for 2D-to-3D conversion. This latter approach exploits the ability of the human visual system to combine reduced disparity information that are located mainly at edges and object boundaries with pictorial depth cues to produce an enhanced sensation of depth over 2D images.