Method of Three-Dimensional and Multiview Video Coding Using a Disparity Vector

摘要

A method and apparatus for a three-dimensional or multi-view video encoding or decoding system are disclosed, where a three-dimensional coding tool relying on a disparity vector are adaptively applied depending on whether the inter-view reference picture pointed by the disparity vector is in the reference list associated with the current slice. The three-dimensional coding tool may correspond to the Inter-View Motion Prediction (IVMP) or View Synthesis Prediction (VSP). If the inter-view reference picture pointed by the DV is not in the current reference list associated with the current slice, the selected three-dimensional coding tool is disabled for the current block. If the inter-view reference picture pointed by the DV is in the current reference list associated with the current slice, the selected three-dimensional coding tool can be applied to the current block.

主权项

1. A method of video coding for a three-dimensional or multi-view video encoding or decoding system, the method comprising:
receiving input data associated with a current block in a dependent view; determining whether an inter-view reference picture pointed by a disparity vector (DV) is in a current reference list associated with a current slice containing the current block for one or more three-dimensional coding tools that utilize the DV; if the inter-view reference picture pointed by the DV is not in the current reference list associated with the current slice, disabling said one or more three-dimensional coding tools for the current block; and if the inter-view reference picture pointed by the DV is in the current reference list associated with the current block, enabling said one or more three-dimensional coding tools for the current block.