László Lois, Tamás Lustyik, Bálint Daróczy, Tibor Agócs, Tibor Balogh
Multiview Video Presentation and Encoding
In this paper the multi-view video encoding and presentation is introduced. In the first part the current multi-view video displays and systems are described. After then we review the developed multi-view video encoding algorithms and the related computer graphics tools. Based the OpenGL system, we show a Depth Image-base Representation (DIBR) method to render an image from several existing reference pictures in a multi-view environment. We also present a new method to encode the depth image efficiently and build a whole multi-view encoding system where the images in the reference views are encoded by using MPEG-4 AVC and the other images are rendered by DIBR. We compare the distortion of this hybrid algorithm and a standard video encoding method in order to establish new multi-view video formats for Holografika’s holographic display. Finally we give some further research objectives to complete the developed hybrid method.