ADAPTIVE OVERLAPPED BLOCK MOTION COMPENSATION

• US 20130128974A1
• Filed: 11/15/2012
• Published: 05/23/2013
• Est. Priority Date: 11/18/2011
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Chien, Wei-Jung, Chong, In Suk, Guo, Liwei, Karczewicz, Marta
• Granted Patent: US 9,883,203 B2
• Time in Patent Office: Days
• US Class Current: 375/240.15
• CPC Class Codes: H04N 19/105 Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties
• CPC Class Codes: H04N 19/139 Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
• CPC Class Codes: H04N 19/17 the unit being an image region, e.g. an object
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/523 with sub-pixel accuracy
• CPC Class Codes: H04N 19/577 Motion compensation with bidirectional frame interpolation, i.e. using B-pictures
• CPC Class Codes: H04N 19/583 Motion compensation with overlapping blocks
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression

Abstract

In general, techniques are described for performing adaptive overlapped block motion compensation when coding video data. A video coding device configured to code video data may implement the techniques. The video coding device may comprise one or more processors configured to determine an adaptation metric from a region of support for one of a first partition and a second partition of one or more blocks of the video data separated by a partition boundary. The region of support may not be the entire one or more blocks of video data. The one or more processors may further be configured to adapt, based on the adaptation metric, application of an overlapped block motion compensation process to determine predicted pixel values near the partition boundary for a first predicted block representative of the first partition of the video block.

Claims

1. A method for coding video data comprising:

determining an adaptation metric from a region of support for one of a first partition and a second partition of one or more blocks of the video data separated by a partition boundary, wherein the region of support is not the entire one or more blocks of video data; and

adapting, based on the adaptation metric, application of an overlapped block motion compensation process to determine predicted pixel values near the partition boundary for a first predicted block representative of the first partition of the video block.

2. The method of claim 1, wherein the region of support comprises one of a frame in which the one or more blocks of the video data resides, a region of the frame in which the block of video data resides, the first partition, the second partition, and a transition region for either the first partition or the second partition that includes a portion of the pixels near the partition boundary in either the first partition or the second partition, a down-sampled version of the first partition, a down-sampled version of the second partition, and a down-sampled version of the transition region for either the first or the second partition.

3. The method of claim 1,

wherein determining the adaptation metric comprises determining a difference between a first predicted pixel value for at least one pixel in the region of support and a second predicted pixel value for the at least one pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel value is predicted from the first motion vector, and

wherein the second predicted pixel value is predicted form the second motion vector.

4. The method of claim 1,

wherein determining the adaptation metric comprises determining an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support.

5. The method of claim 1,

wherein determining the adaptation metric comprises determining an average pixel difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each of the pixels in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel value is predicted form the second motion vector for each of the pixels in the region of support.

6. The method of claim 1, wherein determining the adaptation metric comprises determining an average weighted average of pixels in the region of support.

7. The method of claim 1,

wherein the first partition is associated with a motion vector having a precision,

wherein determining the adaptation metric comprises determining the precision of the motion vector,

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises selectively disabling application of the overlapped block motion compensation process when the precision of the motion vector is less than a threshold precision, and

wherein the threshold precision comprises one of integer precision, half pel precision, and quarter pel precision.

8. The method of claim 1,

wherein the one or more blocks of video data are each associated with a block size,

wherein determining the adaptation metric comprises determining the block size of at least one of the one or more blocks of video data, and

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises selectively disabling application of the overlapped block motion compensation process when the determined block size is less than a block size threshold, and

wherein the determined block size comprises one of a video block size and a transform block size identifying a size of a transform applied to the first partition of the block.

9. The method of claim 1,

wherein the first partition is associated with a prediction direction,

wherein determining the adaptation metric comprises determining the prediction direction, and

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises selectively disabling application of the overlapped block motion compensation process to the second partition when the prediction direction of the first partition is bi-directional.

10. The method of claim 1,

wherein the first partition is associated with a first prediction direction,

wherein the second partition is associated with a second prediction direction,

wherein determining the adaptation metric comprises determining the first prediction direction and the second prediction direction, and

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises:

selectively disabling application of the overlapped block motion compensation process to the first partition when the first prediction direction is bi-directional and the second prediction direction is uni-directional; and

selectively enabling application of the overlapped block motion compensation process to the first partition when the first prediction direction is uni-directional and the second prediction direction is bi-directional.

11. The method of claim 1,

wherein the first partition is associated with a motion vector having a precision,

wherein determining the adaptation metric comprises determining the precision of the motion vector,

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises reducing the precision of the motion vector when precision of the motion vector is fractional to adapt the application of the overlapped block motion compensation process to use the reduced precision motion vector in place of the motion vector associated with the first partition.

12. The method of claim 1,

wherein the first partition is associated with a prediction direction,

wherein determining the adaptation metric comprises determining the prediction direction, and

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises selecting only one of two motion vectors associated with the first partition when the determined prediction direction is bi-directional to adapt application of the overlapped block motion compensation process to use only the selected one of the two motion vectors in place of the two motion vectors.

13. The method of claim 1,

wherein the region of support comprises one or more of a portion of a transition region for the first partition that includes a portion of the pixels near the partition boundary in the first partition, a down-sampled version of the transition region for the first partition, and the first partition,

wherein determining the adaptation metric comprises determining an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support,

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support,

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises selectively disabling application of the overlapped block motion compensation process when the determined absolute difference is greater than a difference threshold.

14. A video coding device configured to code video data, the video coding device comprising:

one or more processors configured to determine an adaptation metric from a region of support for one of a first partition and a second partition of one or more blocks of the video data separated by a partition boundary, wherein the region of support is not the entire one or more blocks of video data, and adapt, based on the adaptation metric, application of an overlapped block motion compensation process to determine predicted pixel values near the partition boundary for a first predicted block representative of the first partition of the video block.

15. The video coding device of claim 14, wherein the region of support comprises one of a frame in which the one or more blocks of the video data resides, a region of the frame in which the block of video data resides, the first partition, the second partition, and a transition region for either the first partition or the second partition that includes a portion of the pixels near the partition boundary in either the first partition or the second partition, a down-sampled version of the first partition, a down-sampled version of the second partition, and a down-sampled version of the transition region for either the first or the second partition.

16. The video coding device of claim 14,

wherein the one or more processors are configured to, when determining the adaptation metric, determine a difference between a first predicted pixel value for at least one pixel in the region of support and a second predicted pixel value for the at least one pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel value is predicted from the first motion vector, and

wherein the second predicted pixel value is predicted form the second motion vector.

17. The video coding device of claim 14,

wherein the one or more processors are configured to, when determining the adaptation metric, determine an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support.

18. The video coding device of claim 14,

wherein the one or more processors are configured to, when determining the adaptation metric, determine an average pixel difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each of the pixels in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel value is predicted form the second motion vector for each of the pixels in the region of support.

19. The video coding device of claim 14, wherein the one or more processors are configured to, when determining the adaptation metric, determining an average weighted average of pixels in the region of support.

20. The video coding device of claim 14,

wherein the first partition is associated with a motion vector having a precision,

wherein the one or more processors are configured to, when determining the adaptation metric, determine the precision of the motion vector,

wherein the one or more processors are configured to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process when the precision of the motion vector is less than a threshold precision, and

wherein the threshold precision comprises one of integer precision, half pel precision, and quarter pel precision.

21. The video coding device of claim 14,

wherein the one or more blocks of video data are each associated with a block size,

wherein the one or more processors are configured to, when determining the adaptation metric, determine the block size of at least one of the one or more blocks of video data, and

wherein the one or more processors are configured to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process when the determined block size is less than a block size threshold, and

wherein the determined block size comprises one of a video block size and a transform block size identifying a size of a transform applied to the first partition of the block.

22. The video coding device of claim 14,

wherein the first partition is associated with a prediction direction,

wherein the one or more processors are configured to, when determining the adaptation metric, determine the prediction direction, and

wherein the one or more processors are configured to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process to the second partition when the prediction direction of the first partition is bi-directional.

23. The video coding device of claim 14,

wherein the first partition is associated with a first prediction direction,

wherein the second partition is associated with a second prediction direction,

wherein the one or more processors are configured to, when determining the adaptation metric, determine the first prediction direction and the second prediction direction, and

wherein the one or more processors are configured to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process to the first partition when the first prediction direction is bi-directional and the second prediction direction is uni-directional and selectively enable application of the overlapped block motion compensation process to the first partition when the first prediction direction is uni-directional and the second prediction direction is bi-directional.

24. The video coding device of claim 14,

wherein the first partition is associated with a motion vector having a precision,

wherein the one or more processors are configured to, when determining the adaptation metric, determine the precision of the motion vector,

wherein the one or more processors are configured to, when adapting application of the overlapped block motion compensation process, reduce the precision of the motion vector when precision of the motion vector is fractional to adapt the application of the overlapped block motion compensation process to use the reduced precision motion vector in place of the motion vector associated with the first partition.

25. The video coding device of claim 14,

wherein the first partition is associated with a prediction direction,

wherein the one or more processors are configured to, when determining the adaptation metric, determine the prediction direction, and

wherein the one or more processors are configured to, when adapting application of the overlapped block motion compensation process, select only one of two motion vectors associated with the first partition when the determined prediction direction is bi-directional to adapt application of the overlapped block motion compensation process to use only the selected one of the two motion vectors in place of the two motion vectors.

26. The video coding device of claim 14,

wherein the region of support comprises one or more of a portion of a transition region for the first partition that includes a portion of the pixels near the partition boundary in the first partition, a down-sampled version of the transition region for the first partition, and the first partition,

wherein the one or more processors are configured to, when determining the adaptation metric, determine an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support,

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support,

wherein adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises selectively disabling application of the overlapped block motion compensation process when the determined absolute difference is greater than a difference threshold.

27. A video coding device configured to code video data, the video coding device comprising:

means for determining an adaptation metric from a region of support for one of a first partition and a second partition of one or more blocks of the video data separated by a partition boundary, wherein the region of support is not the entire one or more blocks of video data; and

means for adapting, based on the adaptation metric, application of an overlapped block motion compensation process to determine predicted pixel values near the partition boundary for a first predicted block representative of the first partition of the video block.

28. The video coding device of claim 27, wherein the region of support comprises one of a frame in which the one or more blocks of the video data resides, a region of the frame in which the block of video data resides, the first partition, the second partition, and a transition region for either the first partition or the second partition that includes a portion of the pixels near the partition boundary in either the first partition or the second partition, a down-sampled version of the first partition, a down-sampled version of the second partition, and a down-sampled version of the transition region for either the first or the second partition.

29. The video coding device of claim 27,

wherein the means for determining the adaptation metric comprises means for determining a difference between a first predicted pixel value for at least one pixel in the region of support and a second predicted pixel value for the at least one pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel value is predicted from the first motion vector, and

wherein the second predicted pixel value is predicted form the second motion vector.

30. The video coding device of claim 27,

wherein the means for determining the adaptation metric comprises means for determining an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support.

31. The video coding device of claim 27,

wherein the means for determining the adaptation metric comprises means for determining an average pixel difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each of the pixels in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel value is predicted form the second motion vector for each of the pixels in the region of support.

32. The video coding device of claim 27,

wherein the first partition is associated with a motion vector having a precision,

wherein the means for determining the adaptation metric comprises means for determining the precision of the motion vector,

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises means for selectively disabling application of the overlapped block motion compensation process when the precision of the motion vector is less than a threshold precision, and

wherein the threshold precision comprises one of integer precision, half pel precision, and quarter pel precision.

33. The video coding device of claim 27,

wherein the one or more blocks of video data are each associated with a block size,

wherein the means for determining the adaptation metric comprises means for determining the block size of at least one of the one or more blocks of video data, and

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises means for selectively disabling application of the overlapped block motion compensation process when the determined block size is less than a block size threshold, and

wherein the determined block size comprises one of a video block size and a transform block size identifying a size of a transform applied to the first partition of the block.

34. The video coding device of claim 27,

wherein the first partition is associated with a prediction direction,

wherein the means for determining the adaptation metric comprises means for determining the prediction direction, and

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises means for selectively disabling application of the overlapped block motion compensation process to the second partition when the prediction direction of the first partition is bi-directional.

35. The video coding device of claim 27,

wherein the first partition is associated with a first prediction direction,

wherein the second partition is associated with a second prediction direction,

wherein the means for determining the adaptation metric comprises means for determining the first prediction direction and the second prediction direction, and

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises:

means for selectively disabling application of the overlapped block motion compensation process to the first partition when the first prediction direction is bi-directional and the second prediction direction is uni-directional; and

means for selectively enabling application of the overlapped block motion compensation process to the first partition when the first prediction direction is uni-directional and the second prediction direction is bi-directional.

36. The video coding device of claim 27,

wherein the first partition is associated with a motion vector having a precision,

wherein the means for determining the adaptation metric comprises means for determining the precision of the motion vector,

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises means for reducing the precision of the motion vector when precision of the motion vector is fractional to adapt the application of the overlapped block motion compensation process to use the reduced precision motion vector in place of the motion vector associated with the first partition.

37. The video coding device of claim 27,

wherein the first partition is associated with a prediction direction,

wherein the means for determining the adaptation metric comprises means for determining the prediction direction, and

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises means for selecting only one of two motion vectors associated with the first partition when the determined prediction direction is bi-directional to adapt application of the overlapped block motion compensation process to use only the selected one of the two motion vectors in place of the two motion vectors.

38. The video coding device of claim 27,

wherein the region of support comprises one or more of a portion of a transition region for the first partition that includes a portion of the pixels near the partition boundary in the first partition, a down-sampled version of the transition region for the first partition, and the first partition,

wherein the means for determining the adaptation metric comprises means for determining an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support,

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support,

wherein the means for adapting, based on the adaptation metric, application of the overlapped block motion compensation process comprises means for selectively disabling application of the overlapped block motion compensation process when the determined absolute difference is greater than a difference threshold.

39. A computer program product comprising a computer-readable storage medium having stored thereon instructions that, when executed, cause one or more processors of a device for coding video data to:

determine an adaptation metric from a region of support for one of a first partition and a second partition of one or more blocks of the video data separated by a partition boundary, wherein the region of support is not the entire one or more blocks of video data; and

adapt, based on the adaptation metric, application of an overlapped block motion compensation process to determine predicted pixel values near the partition boundary for a first predicted block representative of the first partition of the video block.

40. The computer program product of claim 39, wherein the region of support comprises one of a frame in which the one or more blocks of the video data resides, a region of the frame in which the block of video data resides, the first partition, the second partition, and a transition region for either the first partition or the second partition that includes a portion of the pixels near the partition boundary in either the first partition or the second partition, a down-sampled version of the first partition, a down-sampled version of the second partition, and a down-sampled version of the transition region for either the first or the second partition.

41. The computer program product of claim 39,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine a difference between a first predicted pixel value for at least one pixel in the region of support and a second predicted pixel value for the at least one pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel value is predicted from the first motion vector, and

wherein the second predicted pixel value is predicted form the second motion vector.

42. The computer program product of claim 39,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support.

43. The computer program product of claim 39,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine an average pixel difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each of the pixels in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support, and

wherein the second predicted pixel value is predicted form the second motion vector for each of the pixels in the region of support.

44. The computer program product of claim 39,

wherein the first partition is associated with a motion vector having a precision,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine the precision of the motion vector,

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process when the precision of the motion vector is less than a threshold precision, and

wherein the threshold precision comprises one of integer precision, half pel precision, and quarter pel precision.

45. The computer program product of claim 39,

wherein the one or more blocks of video data are each associated with a block size,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine the block size of at least one of the one or more blocks of video data, and

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process when the determined block size is less than a block size threshold, and

wherein the determined block size comprises one of a video block size and a transform block size identifying a size of a transform applied to the first partition of the block.

46. The computer program product of claim 39,

wherein the first partition is associated with a prediction direction,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine the prediction direction, and

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process to the second partition when the prediction direction of the first partition is bi-directional.

47. The computer program product of claim 39,

wherein the first partition is associated with a first prediction direction,

wherein the second partition is associated with a second prediction direction,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine the first prediction direction and the second prediction direction, and

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process:

selectively disable application of the overlapped block motion compensation process to the first partition when the first prediction direction is bi-directional and the second prediction direction is uni-directional; and

selectively enable application of the overlapped block motion compensation process to the first partition when the first prediction direction is uni-directional and the second prediction direction is bi-directional.

48. The computer program product of claim 39,

wherein the first partition is associated with a motion vector having a precision,

wherein the instructions, when executed, cause the one or more processors to, when determine the adaptation metric, determine the precision of the motion vector,

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process, reduce the precision of the motion vector when precision of the motion vector is fractional to adapt the application of the overlapped block motion compensation process to use the reduced precision motion vector in place of the motion vector associated with the first partition.

49. The computer program product of claim 39,

wherein the first partition is associated with a prediction direction,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine the prediction direction, and

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process, select only one of two motion vectors associated with the first partition when the determined prediction direction is bi-directional to adapt application of the overlapped block motion compensation process to use only the selected one of the two motion vectors in place of the two motion vectors.

50. The computer program product of claim 39,

wherein the region of support comprises one or more of a portion of a transition region for the first partition that includes a portion of the pixels near the partition boundary in the first partition, a down-sampled version of the transition region for the first partition, and the first partition,

wherein the instructions, when executed, cause the one or more processors to, when determining the adaptation metric, determine an absolute difference between first predicted pixel values for each pixel in the region of support and second predicted pixel values for each pixel in the region of support,

wherein the first partition is associated with a first motion vector referencing a first block of a first reference frame,

wherein the second partition is associated with a second motion vector referencing a second block of a second reference frame,

wherein the first predicted pixel values are predicted from the first motion vector for each of the pixels in the region of support,

wherein the second predicted pixel values are predicted form the second motion vector for each of the pixels in the region of support,

wherein the instructions, when executed, cause the one or more processors to, when adapting application of the overlapped block motion compensation process, selectively disable application of the overlapped block motion compensation process when the determined absolute difference is greater than a difference threshold.