Texture-based depth frame interpolation for precise 2D-to-3D conversion

In general, if the 3D videos are represented by texture and its depth frames, the 3D multiview contents can be effectively produced by depth-image-based rendering (DIBR). In recent years, many researches are proposed to deal with the estimation of the depth map by using stereo images. For many 2D movies, the traditional depth cue methods are limited by specified scenery and achieve poor depth quality. In this paper, we proposed a precise depth map interpolation algorithm to estimate depth maps from two known depth maps as the depth keyframes and the color texture frames. After proper computation of superpixels, the proposed depth frame interpolation system contains texture-based depth estimation, error compensation, noise elimination, and forward/backward depth map merging steps. Simulations show that the proposed system can obtain high quality depth frame interpolation.