Method for encoding and decoding video including plurality of layers
A method for decoding a video including a plurality of views, according to one embodiment of the present invention, comprises the steps of: configuring a base merge motion candidate list by using motion information of neighboring blocks and a time correspondence block of a current block; configuring an extended merge motion information list by using motion information of a depth information map and a video view different from the current block; and determining whether neighboring block motion information contained in the base merge motion candidate list is derived through view synthesis prediction.
1. A processor implemented method of decoding a video comprising views, the method comprising:
constructing a basic merge motion candidate list for an independent view using motion information on a spatial neighboring block to a current block and motion information on a collocated block in a temporal reference picture;
determining whether a picture to which the current block belongs is a depth map information picture or a dependent view picture;
in response to the determining, constructing an extended merge motion candidate list based on the basic merge motion candidate list using information on a view identifier of the dependent view or information on the depth map; and
decoding, based on the determination, the video comprising the views using the extended merge motion candidate list or the basic merge motion candidate list,
wherein the extended merge motion candidate list comprises any one or any combination of two or more of information on a block of a different view corresponding to the current block, view synthesis prediction information, and inter-view disparity information.
2. The method of claim 1 , wherein the constructing of the extended merge motion candidate list further comprises adding motion information (candT) on a collocated picture block with the current block when the depth map is encoded, and the adding of candT comprises determining availability of candT and adding candT to the extended merge motion candidate list as a first candidate in response to candT being available.
3. The method of claim 1 , wherein the constructing of the extended merge motion candidate list further comprises adding motion information (candIVT) on a neighboring view block corresponding to the current block, and the adding of candIVT comprises determining availability of candIVT and adding candIVT to the extended merge motion candidate list as a first candidate in response to candIVT being available.
4. The method of claim 1 , wherein the constructing of the extended merge motion candidate list further comprises adding inter-view disparity information (candIVD), and the adding of candIVD comprises determining availability of candIVD and adding candIVD to the extended merge motion candidate list after the motion information on the neighboring block in response to candIVD being available.
5. The method of claim 1 , wherein the constructing of the extended merge motion candidate list further comprises adding motion information (candVSP) by a view synthesis prediction and the adding of candVSP comprises determining availability of candVSP and adding candVSP to the extended merge motion candidate list after the motion information on the neighboring block in response to candVSP being available.
6. The method of claim 1 , further comprising determining whether motion information on the neighboring block comprised in the basic merge motion candidate list is generated using view synthesis prediction.
7. An apparatus for decoding a video comprising views, the apparatus comprising:
a basic merge motion candidate constructor configured to construct a basic merge motion candidate list using motion information on a spatial neighboring block to a current block and motion information on a collocated block in a temporal reference picture; and
an extended merge motion list constructor configured to construct an extended merge motion candidate list based on the basic merge motion candidate list using motion information on a view picture different from the current block or motion information on a depth map,
wherein the extended merge motion list constructor is invoked to construct the extended merge motion candidate list according to a determination that a picture to which the current block belongs is a depth map information picture or a dependent view picture, and
wherein the extended merge motion candidate list comprises any one or any combination of two or more of information on a block of a different view corresponding to the current block, view synthesis prediction information, and inter-view disparity information; and
a video decoder configured to decode, based on the determination, the video comprising the views using the extended merge motion candidate list or the basic merge motion candidate list.
8. The apparatus of claim 7 , wherein the extended merge motion list constructor is further configured to add motion information (candT) on a collocated picture block with the current block in response to the depth map being encoded, and in response to adding candT, the extended merge motion list construction module is further configured to determine availability of candT and add candT to the extended merge motion candidate list as a first candidate in response to candT being available.
9. The apparatus of claim 7 , wherein the extended merge motion list constructor is further configured to add motion information (candIVT) on a neighboring view block corresponding to the current block, and in response to adding candIVT, the extended merge motion list construction module is further configured to determine availability of candIVT and add candIVT to the extended merge motion candidate list as a first candidate in response to candIVT being available.
10. The apparatus of claim 7 , wherein the extended merge motion list constructor is further configured to add inter-view disparity information (candIVD), and when adding candIVD, the extended merge motion list construction module determines availability of candIVD and add candIVD to the extended merge motion candidate list after the motion information on the neighboring block in response to candIVD being available.
11. The apparatus of claim 7 , wherein the extended merge motion list constructor is further configured to add motion information (candVSP) by a view synthesis prediction, and when adding candVSP the extended merge motion list construction module determines availability of candVSP and add candVSP to the extended merge motion candidate list after the motion information on the neighboring block in response to candVSP being available.
12. The apparatus of claim 7 , wherein the extended merge motion list constructor is further configured to determine whether motion information on the neighboring block comprised in the basic merge motion candidate list is generated using view synthesis prediction.