| 摘要 |
<p>In one embodiment, the method includes parsing data from a data stream for the first picture layer into a sequence of data blocks on a cycle-by-cycle basis such that at least one data block earlier in the sequence is skipped during a cycle if a data block later in the sequence includes an empty data location closer to DC components than in the earlier data block. A motion vector pointing to a reference block for at least one of the data blocks is generated based on motion vector information for a block in a second picture layer and motion vector difference information associated with the data block. The second picture layer represents lower quality pictures than pictures represented by the first picture layer, and the block of the second picture layer is temporally associated with the data block in the first picture layer. An image block is reconstructed based on the data block and the reference block.</p> |
