| A Method and Apparatus for Multi-View Video Coding 
Description Video coding technologies have matured sufficiently to make possible a new generation of video applications. This invention is related to the field of video coding and, more particularly, video coding compression for multi view video. Multi-view video coding (MVC) has begun receiving significant attention in the industry and standardization is underway in the MPEG committee. The biggest challenge is developing compression technologies that can exploit the redundancies in multiple views to achieve high compression ratio. The compression algorithm has to support fast decompression and play the selected views at the receiver. A more efficient mechanism for performing MVC is needed. This invention addresses this requirement through an apparatus which consists of a multi-view video encoder and video decoder, which uses a novel prediction structure for MVC. This new MVC structure is based on a hypercube configuration, resulting in minimal dependencies among the views and at the same time allows exploitation of the redundancies efficiently. The method uses a very compact notation to identify the views and dependencies thus resulting in the transmission of fewer bits to the receiver. The apparatus can be designed using hardware or software components, or as a combination of both. Potential Applications The goal of MVC is to allow coding of multiple camera views such that the user has the freedom of choosing the view point. Free viewpoint TV (FTV) is considered a key application. This technology also has applications in security and surveillance. State of Technology Development and Commercialization Demonstrations available
(c) Florida Atlantic University All rights reserved.
Updated August 9, 2008 | 
