Affine model for disparity estimation in multiview sequence coding

One major obstacle for the application of multiview sequences, including 3DTV, is the extremely large amount of data associated with a multiview sequence. To enable the storage or transmission of a multiview sequence at a reasonable cost, in this paper, we present a disparity estimation method based on affine model to overcome this problem. By modeling the disparity function on each patch of the multiview image, the proposed technique can improve the accuracy of the disparity estimation and gain the compressing performance efficiently.