A novel depth coding scheme for multiview video plus depth compression

Efficient compression of multi-view images and videos is an open and interesting research issue that has been attracting the attention of both academic and industrial world during the past several years. Multiview video plus depth (MVD) is a new 3D video format that would support 3D television (3DTV) and free viewpoint television (FTV). Depth-images are a 2D representation of the 3D scene surface and for 3DTV and FTV applications the depth information of MVD data is used for rendering virtual intermediate views of the scene. In this paper, we propose a novel coding scheme that is adapted to the special characteristics of depth-images, individual coding for color and depth data: the conventional multiview coding approach (MVC) is used for color videos and multidimensional multiscale parser (MMP) algorithm for depth maps coding. Experimental results show that the presented scheme achieves significant coding gains at high rate while compared with the joint multiview video model and Platelet depth coding.