| 摘要 |
Methods include receiving a multi-view video stream including a random access picture including a random access slice referencing only a slice corresponding to a same time and a different view of the picture; obtaining identification information representing the stream including initialization information of a reference picture list for the slice; obtaining the initialization information based on the identification information, the initialization information representing view relationships between several views and including view number information and view identification information for the views; obtaining type information indicating whether the slice is predicted from one or two reference pictures; initializing the list using the view number and view identification information according to the type information; determining a prediction value of a macroblock in the picture based on the initialized list; and decoding the macroblock using the prediction value, the initialization information being obtained based on a value indicating decoding order between the views.
|
