Hierarchial video code block merging using depth-dependent threshold for block merger
A computer executing: a vector search process for determining motion information for each of the smallest blocks, which are the blocks in a predetermined deepest layer; a block size determination process for determining the block size, which is the unit in which the motion information is to be held; and a block integration process for determining whether to integrate mutually adjacent sub-blocks with a single block having a single piece of motion information, based on integration conditions including at least the motion information of the sub-blocks which are the blocks in a layer one layer deeper than the layer currently being considered; wherein the block size determination process determines the block size by carrying out the block integration process in layers, in a sequence from the blocks in a layer one level shallower than the smallest blocks, to the largest blocks, which are the blocks in a predetermined shallowest layer.
1. A non-transitory computer-readable program recording medium recording a program for a video coding method for estimating motion, the program making a computer execute:
vector search processing of obtaining motion information of each of smallest blocks that are blocks belonging to a predefined deepest layer;
block size determination processing of determining a block size that is a unit for retaining motion information; and
block merging processing of, based on a merging condition that includes at least motion information of sub-blocks that are blocks belonging to a layer one layer deeper than a layer on which attention is currently focused, determining whether one or a plurality of sub-blocks that are adjacent to each other are merged into a block including a piece of motion information;
wherein the block size determination processing determines a block size by means of executing the block merging processing in a hierarchical manner in the order from blocks belonging to a layer one layer shallower than the smallest blocks to a largest block that is a block belonging to a predefined shallowest layer,
wherein the merging condition includes a condition that is defined based on a comparison between a difference in motion information between adjacent sub-blocks and a threshold value related to the difference, and
wherein the merging condition includes a depth of a layer on which attention is currently focused, and
the program further makes the computer execute processing of defining a threshold value related to the difference depending on a depth of a layer on which attention is currently focused.
2. The non-transitory computer-readable program recording medium according to claim 1 , wherein
the merging condition includes another piece of information of motion information of sub-blocks; and
the program further makes the computer execute merging condition determination processing of determining the merging condition at a layer on which attention is currently focused based on the another piece of information.
3. The non-transitory computer-readable program recording medium according to claim 2 , wherein
the another piece of information is a depth of a layer on which attention is currently focused.
4. The non-transitory computer-readable program recording medium according to claim 2 , wherein
the another piece of information is information relating to a result from the block merging processing at a layer deeper than a layer on which attention is currently focused.
5. The non-transitory computer-readable program recording medium according to claim 1 , the program further making the computer execute:
processing of, based on a result from the block merging processing at a layer deeper than a layer on which attention is currently focused, determining whether execution of the block merging processing at a layer on which attention is currently focused is needed.
6. A block size determination method for a video coding method for estimating motion, the block size determination method comprising:
obtaining motion information of each of smallest blocks that are blocks belonging to a predefined deepest layer;
determining a block size that is a unit for retaining motion information; based on a merging condition that includes at least motion information of sub-blocks that are blocks belonging to a layer one layer deeper than a layer on which attention is currently focused, determining whether one or a plurality of sub-blocks that are adjacent to each other are merged into a block including a piece of motion information;
wherein the block size is determined by means of performing the determination in a hierarchical manner in the order from blocks belonging to a layer one layer shallower than the smallest blocks to a largest block that is a block belonging to a predefined shallowest layer,
wherein the merging condition includes a condition that is defined based on a comparison between a difference in motion information between adjacent sub-blocks and a threshold value related to the difference, and
wherein the merging condition includes a depth of a layer on which attention is currently focused, and
defining a threshold value related to the difference depending on a depth of a layer on which attention is currently focused.
7. The block size determination method according to claim 6 , wherein
the merging condition includes another piece of information of motion information of sub-blocks, and
the block size determination method further comprises determining the merging condition at a layer on which attention is currently focused based on the another piece of information.
8. The block size determination method according to claim 7 , wherein
the another piece of information is a depth of a layer on which attention is currently focused.