Multi-view video composition with computer graphics

Multi-view video has recently gained much attraction from academic and commercial fields because it can deliver the immersive viewing of natural scenes. This paper presents a composition method of generating composite views combined with graphic objects. First we generate virtual views between multi-view cameras using depth and texture images of the input videos. Then we mix graphic objects to the generated view images. A distinctive feature of our approach is that we use an identical coordinate system for camera, virtual, and graphics views. The reason for using the same system for all types of cameras is the ability of full interactions between real scene and graphic objects. Another merit is that the occlusion between them is handled automatically by a graphics engine because z values of real scene and graphic objects are stored in the same z buffer. We present experimental results that validate our proposed method and show that graphic objects could become the inalienable part of the multi-view video. For the validation of our method we used multi-view sequences where a graphic object is mixed into camera and virtual images.