An effective hybrid depth-perception algorithm for 2D-to-3D conversion in 3D display systems

3D-TV plays an important revolution in consumer electronics. However, depth map generation is a critical technology for 2D to 3D conversion. In this paper, we propose a hybrid depth-perception algorithm for 2D-to-3D conversion in 3D display systems. The algorithm consists of three parts: moving object of border extraction, vanish line detection, and smoothing of the depth map. From the experimental results by using 3D active glasses, the quality of depth perception is good enough, hence the human eye can't find noticeable differences. Furthermore, the proposed algorithm is also suitable for VLSI design.