Adaptive transform size selection for geometric motion partitioning

Patent metadata

Field	Value
Publication number	US-2017201770-A1
Application number	US-201715470556-A
Country	US
Kind code	A1
Filing date	Mar 27, 2017
Priority date	Feb 18, 2010
Publication date	Jul 13, 2017
Grant date	—

Abstract

In one example, an apparatus includes a video encoder configured to partition a block of video data into a first geometric partition and a second geometric partition using a geometric motion partition line, wherein the block comprises N×N pixels, divide the block of video data into four equally-sized, non-overlapping (N/2)×(N/2) sub-blocks, and encode at least one of the sub-blocks through which the geometric motion partition line passes using a transform size smaller than (N/2)×(N/2). The video encoder may determine transform sizes for the sub-blocks based on whether the geometric motion partition line passes through the sub-blocks. In one example, a video decoder may inverse transform the sub-blocks, and may determine transform sizes for the sub-blocks based on whether the geometric motion partition line passes through the sub-blocks.

First claim

1 . A method of encoding video data, the method comprising:
partitioning a block of video data into a first partition and a second partition using a geometric motion partition line; determining a first motion vector for the first partition and a second motion vector for the second partition; encoding the first motion vector based on a first motion predictor selected from motion vectors for blocks neighboring the first partition; encoding the second motion vector based on a second motion predictor selected from motion vectors for blocks neighboring the second partition, wherein the blocks neighboring the second partition are determined independently of the blocks neighboring the first partition; and outputting the encoded first and second motion vectors.

2 . The method of claim 1 , further comprising:
determining a set of blocks that neighbor the partitioned block; determining a first subset of the set of blocks comprising blocks that neighbor the first partition; and determining a second subset of the set of blocks comprising blocks that neighbor the second partition, independently of determining the first subset.

3 . The method of claim 2 , further comprising:
determining the first motion predictor by calculating the median of motion vectors for blocks of the first subset; and determining the second motion predictor by calculating the median of motion vectors for blocks of the second subset.

4 . The method of claim 2 , wherein the set of blocks that neighbor the partitioned block comprises more than four blocks.

5 . The method of claim 2 , wherein the set of blocks that neighbor the partitioned block comprises one or more of the group consisting of: blocks above the partitioned block, blocks to the left of the partitioned block, sub-blocks of a block to the left of the partitioned block, and sub-blocks of a block above the partitioned block.

6 . The method of claim 1 , wherein the geometric motion partition line partitions the block of video data into non-rectangular partitions such that the geometric motion partition line intersects one horizontal boundary of the block of video data and one vertical boundary of the block of video data.

7 . An apparatus for encoding video data, the apparatus comprising a video encoder configured to partition a block of video data into a first partition and a second partition using a geometric motion partition line, determine a first motion vector for the first partition and a second motion vector for the second partition, encode the first motion vector based on a first motion predictor selected from motion vectors for blocks neighboring the first partition, encode the second motion vector based on a second motion predictor selected from motion vectors for blocks neighboring the second partition, wherein the blocks neighboring the second partition are determined independently of the blocks neighboring the first partition, and output the encoded first and second motion vectors.

8 . The apparatus of claim 7 , wherein the video encoder is configured to determine a set of blocks that neighbor the partitioned block, determine a first subset of the set of blocks comprising blocks that neighbor the first partition, and determine a second subset of the set of blocks comprising blocks that neighbor the second partition, independently of determining the first subset.

9 . The apparatus of claim 8 , wherein the video encoder is configured to determine the first motion predictor by calculating the median of motion vectors for blocks of the first subset, and to determine the second motion predictor by calculating the median of motion vectors for blocks of the second subset.

10 . The apparatus of claim 8 , wherein the set of blocks that neighbor the partitioned block comprises more than four blocks.

11 . The apparatus of claim 8 , wherein the set of blocks that neighbor the partitioned block comprises one or more of the group consisting of: blocks above the partitioned block, blocks to the left of the partitioned block, sub-blocks of a block to the left of the partitioned block, and sub-blocks of a block above the partitioned block.

12 . The apparatus of claim 7 , wherein the geometric motion partition line partitions the block of video data into non-rectangular partitions such that the geometric motion partition line intersects one horizontal boundary of the block of video data and one vertical boundary of the block of video data

13 . An apparatus for encoding video data, the apparatus comprising:
means for partitioning a block of video data into a first partition and a second partition using a geometric motion partition line; means for determining a first motion vector for the first partition and a second motion vector for the second partition; means for encoding the first motion vector based on a first motion predictor selected from motion vectors for blocks neighboring the first partition; means for encoding the second motion vector based on a second motion predictor selected from motion vectors for blocks neighboring the second partition, wherein the blocks neighboring the second partition are determined independently of the blocks neighboring the first partition; and means for outputting the encoded first and second motion vectors.

14 . The apparatus of claim 13 , further comprising:
means for determining a set of blocks that neighbor the partitioned block; means for determining a first subset of the set of blocks comprising blocks that neighbor the first partition; and means for determining a second subset of the set of blocks comprising blocks that neighbor the second partition, independently of determining the first subset.

15 . The apparatus of claim 14 , further comprising:
means for determining the first motion predictor by calculating the median of motion vectors for blocks of the first subset; and means for determining the second motion predictor by calculating the median of motion vectors for blocks of the second subset.

16 . The apparatus of claim 14 , wherein the set of blocks that neighbor the partitioned block comprises more than four blocks.

17 . The apparatus of claim 14 , wherein the set of blocks that neighbor the partitioned block comprises one or more of the group consisting of: blocks above the partitioned block, blocks to the left of the partitioned block, sub-blocks of a block to the left of the partitioned block, and sub-blocks of a block above the partitioned block.

18 . The apparatus of claim 13 , wherein the geometric motion partition line partitions the block of video data into non-rectangular partitions such that the geometric motion partition line intersects one horizontal boundary of the block of video data and one vertical boundary of the block of video data.

19 . A computer program product comprising a non-transitory computer-readable storage medium having stored thereon instructions that, when executed, cause a processor to:
partition a block of video data into a first partition and a second partition using a geometric motion partition line; determine a first motion vector for the first partition and a second motion vector for the second partition; encode the first motion vector based on a first motion predictor selected from motion vectors for blocks neighboring the first partition; encode the second motion vector based on a second motion predictor selected from motion vectors for blocks neighboring the second partition, wherein the blocks neighboring the second partition are determined independently of the blocks neighboring the first

Assignees

• Qualcomm Inc

Inventors

• Chen Peisong
• Wang Xianglin
• Karczewicz Marta

Classifications

• H04N19/122

  Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type · CPC title

• H04N19/119 Primary

  Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks · CPC title

• H04N19/17

  the unit being an image region, e.g. an object · CPC title

• H04N19/174

  the region being a slice, e.g. a line of blocks or a group of blocks · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title