Wavelet Compression of Multiview Video with Disparity-Compensated Lifting

In recent years, there has been an increasing interest in generating free-viewpoint images of a scene from multiple camera views. This will provide viewers with more vivid visual entertainment as well as interactivity. But with the huge amount of date taken by multiple cameras, much higher compression rates are needed to transmit multiview information in acceptable times, e.g. over the Internet or wireless media. In this paper, disparity compensation is incorporated into wavelet coding elegantly by lifting scheme. By this approach, both inter-view and intra-view spatial coherence is fully exploited so as to achieve superior compression efficiency as well as scalability. A scheme that facilitates optimal allocation of bit rates to the different components of a lifted scheme (the L/H images and the disparity vectors) is also proposed.