High accuracy motion vectors for video coding with low encoder and decoder complexity

• US 8,275,041 B2
• Filed: 04/08/2008
• Issued: 09/25/2012
• Est. Priority Date: 04/09/2007
• Status: Active Grant

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Corporation
• Inventors: Ugur, Kemal, Lainema, Jani
• Primary Examiner(s): Yao, Kwang B
• Assistant Examiner(s): NGO, NGUYEN HOANG
• Application Number:
  US12/099,731
• Publication Number:
  US 20080253459A1
• Time in Patent Office: 1,631 Days
• Field of Search: 375/240.16, 375/240.17, 375/240, 375/240.01, 375/240.24, 375/240.08
• US Class Current: 375/240.16
• CPC Class Codes: H04N 19/523 with sub-pixel accuracy
• CPC Class Codes: H04N 19/53 Multi-resolution motion estimation; Hierarchical motion estimation
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

Two-stage interpolation can be provided for frame prediction samples with quarter-pixel and finer accuracy. All samples of quarter-pixel and finer accuracy can use either half or full/integer-pixels in a bi-linear interpolation to allow for the use of higher accuracy motion vectors, such as one-eighth-pixel accuracy motion vectors. The motion vectors can be restricted in a manner such that they are not allowed to point to every possible sub-pixel sample on a sub-pixel grid, but rather a subset of those sub-pixel samples. In addition, the same full/integer and half-pixel samples that can be used to obtain a quarter-pixel sample can also be used to obtain a one-eighth-pixel sample that the quarter-pixel sample is connected to. Hence, for every quarter-pixel sample, a motion vector could point to two additional positions/locations with one-eighth-pixel accuracy, where the two additional positions/locations can be calculated using the same half or integer-pixel samples which were used to obtain the quarter-pixel sample. Therefore, an arbitrary motion vector accuracy can be achieved without the need to perform additional interpolation stages or increase the complexity of a decoder configured to decode motion vector data.

Claims

1. A method of decoding motion vector data encoded in a bitstream, comprising:

determining that a pixel offset syntax element is encoded in the bitstream and that a motion vector has at least quarter-pixel accuracy;

reading the pixel offset syntax element;

utilizing the motion vector having the at least quarter-pixel accuracy to calculate a position of a first sample having the at least quarter-pixel accuracy to which the motion vector points;

utilizing a fractional interpolation structure algorithm to obtain a second sample having a pixel accuracy greater than the first sample;

wherein the second sample is representative of one of at least two additional sample positions associated with the first sample, the at least two additional sample positions each having an accuracy greater than the at least quarter-pixel accuracy; and

wherein the at least two additional sample positions are calculated using at least one of a full-pixel sample position and a half-pixel sample position, and wherein the at least one of the full-pixel sample position and the half-pixel sample position is the same sample position utilized to determine the first sample having the at least quarter-pixel accuracy.

2. The method of claim 1, wherein the pixel offset syntax element comprises a one-eighth-pixel offset syntax element and the second sample comprises a one-eighth-pixel accuracy sample.

3. The method of claim 1, wherein the pixel offset syntax element and the second sample have an accuracy greater than one-eighth-pixel accuracy.

4. The method of claim 1, wherein the pixel offset syntax element is representative of a pixel offset and the pixel offset is indicative of a position of the second sample.

5. The method of claim 4, wherein the pixel offset syntax element has no dependency upon another pixel offset syntax element present in a macroblock header and the pixel offset is not predicted from a neighboring macroblock header.

6. The method of claim 4, wherein the pixel offset is predicted from a neighboring macroblock.

7. The method of claim 4, wherein the pixel offset is predicted based on a motion vector difference coded in the bitstream.

8. The method of claim 4, wherein the pixel offset is indicative of a motion accuracy for at least one a picture, a group of pictures, a macroblock, a group of macroblocks, and a block in relation to the pixel offset.

9. The method of claim 8, wherein the motion accuracy depends on a mode of at least the macroblock.

10. The method of claim 4 further comprising, utilizing an extended fractional interpolation structure algorithm if the offset syntax element is dependent upon another offset syntax element.

11. The method of claim 1, wherein the motion vector does not point to every possible sub-pixel sample on a sub-pixel grid including at least the first sample and the second sample.

12. A computer program product, embodied on a non-transitory computer-readable medium, comprising computer code configured to perform the processes of claim 1.

13. An apparatus, comprising:

a processor; and

a memory unit operatively connected to the processor and including computer coded instructions configured to, with the processor, cause the apparatus to;

determine that a pixel offset syntax element is encoded in a bitstream and that a motion vector encoded in the bitstream has at least quarter-pixel accuracy;

read the pixel offset syntax element;

utilize the motion vector having the at least quarter-pixel accuracy to calculate a position of a first sample having the at least quarter-pixel accuracy to which the motion vector points;

utilize a fractional interpolation structure algorithm to obtain a second sample having a pixel accuracy greater than the first sample;

wherein the second sample is representative of one of at least two additional sample positions associated with the first sample, the at least two additional sample positions each having an accuracy greater than the at least quarter-pixel accuracy; and

wherein the at least two additional sample positions are calculated using at least one of a full-pixel sample position and a half-pixel sample position, and wherein the at least one of the full-pixel sample position and the half-pixel sample position is the same sample position utilized to determine the first sample having the at least quarter-pixel accuracy.

14. The apparatus of claim 13, wherein the pixel offset syntax element comprises a one-eighth-pixel offset syntax element and the second sample comprises a one-eighth-pixel accuracy sample.

15. The apparatus of claim 13, wherein the pixel offset syntax element is representative of a pixel offset and the pixel offset is indicative of a position of the second sample.

16. The apparatus of claim 13, wherein the motion vector is restricted from pointing to every possible sub-pixel sample on a sub-pixel grid including at least the first sample and the second sample.