3D Video Systems
Overview
Stereoscopic video is two-channel video taken from a binocular camera which provides viewers images with depth information. Recently the auto-stereoscopic display becomes possible due to the development of optical and LCD technology which may make stereoscopic video more popular soon. A simplified object based stereoscopic video coding process is shown in Fig. 1.
Traditionally, disparity matching is done for each pair of images. However, we proposed two level hybrid disparity matching method suitable for video pair and also ABKB depth map segmentation algorithm. The processing results of these methods can be used for coding purpose. Some 3D motion estimation improving techniques and other techniques are also explored.
Figure 1. 3D Video Coding
Related Papers
J. Lin and K.K. Parhi, "VLSI Architectures for Stereoscopic Video, Disparity Matching and Object Extraction", Proc. of 2005 IEEE Int. Symposium on Circuits and Systems, pp. 2373-2376, Kobe (Japan), May 2005 Back
