Method and apparatus for parallelizing context selection in video processing

Patent metadata

Field	Value
Publication number	US-10142637-B2
Application number	US-201213415550-A
Country	US
Kind code	B2
Filing date	Mar 8, 2012
Priority date	Mar 8, 2011
Publication date	Nov 27, 2018
Grant date	Nov 27, 2018

Abstract

A method and apparatus for parallel processing of at least two bins relating to at least one of a video and an image. The method includes determining scan type of at least a portion of the at least one of video and an image, analyzing neighboring position of a bin, removing dependencies of context selection based on the scan type and position of location being encoded in a transform, and performing parallel processing of that least two bins.

First claim

What is claimed is:

1. A method for parallel context processing for image decoding, the method comprising:
receiving, by at least one processor of a plurality of parallel processors, an image;
dividing the bins into a plurality of groups of N bins based on a zig-zag scan order that includes a plurality of diagonal scan lines, each of the groups of N bins being a set of N contiguous bins along the zig-zag scan order, wherein N is an integer greater than 1;
generating, by the at least one processor, reduced context bins for bins in a scan representing a significance map of the image, comprising:
removing, by the at least one processor, a context selection dependency directly above a selected bin of the bins, in response to determining that the selected bin is in a leftmost column; and
removing, by the at least one processor, a context selection dependency directly to the left of the selected bin, in response to determining that the selected bin is in a top row; and
performing, by the plurality of parallel processors, parallel context processing for the reduced context bins, to generate a decoded image; and

outputting, by the at least one processor, the decoded image.

2. The method of claim 1 , wherein the scan is at least one of a reverse zig-zag scan, a forward zig-zag scan, a diagonal scan, a sub-block diagonal scan, a horizontal scan, or a vertical scan.

3. The method of claim 1 , wherein generating the reduced context bins for the bins is performed using a mask.

4. The method of claim 1 , wherein generating the reduced context bins for the bins further comprises:
removing a context selection dependency directly above the selected bin of the bins, in response to determining that the selected bin is in a column second from the left; and
removing a context selection dependency directly to the left of the selecting bin, in response to determining that the selected bin is in a row second from the top.

5. The method of claim 4 , wherein generating the reduced context bins for the bins further comprises:
removing a context selection dependency directly left of the selected bin of the bins, in response to determining that the selected bin is in a column second from the left; and
removing a context selection dependency directly to the above the selecting bin, in response to determining that the selected bin is in a row second from the top.

6. The method of claim 1 , further comprising displaying the decoded image.

7. A non-transitory computer readable storage medium storing a program for execution by at least one processor of a plurality of parallel processors, the program including instructions for parallel context processing for image decoding, the instructions to:
receive an image;
divide the bins into a plurality of groups of N bins based on a zig-zag scan order that includes a plurality of diagonal scan lines, each of the groups of N bins being a set of N contiguous bins along the zig-zag scan order, wherein N is an integer greater than 1;
generate reduced context bins for bins in a scan representing a significance map of the image, comprising instructions to:
remove a context selection dependency directly above a selected bin of the bins, in response to determining that the selected bin is in a leftmost column; and
remove a context selection dependency directly to the left of the selected bin, in response to determining that the selected bin is in a top row;

perform parallel context processing for the reduced context bins, to generate a decoded image; and
output the decoded image.

8. The non-transitory computer readable storage medium of claim 7 , wherein the scan type is at least one of a reverse zig-zag scan, a forward zig-zag scan, a diagonal scan, a horizontal scan, or a vertical scan.

9. The non-transitory computer readable storage medium of claim 7 , wherein the instructions to generate the reduced context for the bins comprise instructions to generate the reduced context using a mask.

10. The non-transitory computer readable storage medium of claim 7 , wherein the instructions to generate the reduced context bins for the bins further comprises instructions to:
remove a context selection dependency directly above the selected bin of the bins, in response to determining that the selected bin is in a column second from the left; and
remove a context selection dependency directly to the left of the selecting bin, in response to determining that the selected bin is in a row second from the top.

11. The non-transitory computer readable storage medium of claim 10 , wherein the instructions to generate the reduced context bins for the bins further comprises instructions to:
remove a context selection dependency directly left of the selected bin of the bins, in response to determining that the selected bin is in a column second from the left; and
remove a context selection dependency directly to the above the selecting bin, in response to determining that the selected bin is in a row second from the top.

12. A method for parallel context processing for image encoding, the method comprising:
receiving, by at least one processor of a plurality of parallel processors, an image;
dividing the bins into a plurality of groups of N bins based on a zig-zag scan order that includes a plurality of diagonal scan lines, each of the groups of N bins being a set of N contiguous bins along the zig-zag scan order, wherein N is an integer greater than 1;
generating, by the at least one processor, reduced context bins for bins in a scan representing a significance map of the image, comprising:
removing, by the at least one processor, a context selection dependency directly above a selected bin of the bins, in response to determining that the selected bin is in a leftmost column; and
removing, by the at least one processor, a context selection dependency directly to the left of the selected bin, in response to determining that the selected bin is in a top row;

performing, by the plurality of parallel processors, parallel context processing for the reduced context bins, to generate a decoded image; and
outputting, by the digital processor, the decoded image.

13. The method of claim 12 , wherein receiving the image comprises receiving the image from a video capture device.

14. The method of claim 12 , wherein receiving the image comprises retrieving the image from memory.

15. The method of claim 12 , wherein generating the reduced context bins for the bins further comprises:
removing a context selection dependency directly above the selected bin of the bins, in response to determining that the selected bin is in a column second from the left; and
removing a context selection dependency directly to the left of the selecting bin, in response to determining that the selected bin is in a row second from the top.

16. The method of claim 15 , wherein generating the reduced context bins for the bins further comprises:
removing a context selection dependency directly left of the selected bin of the bins, in response to determining that the selected bin is in a column second from the left; and
removing a context selection dependency directly to the above the selecting bin, in response to determining that the selected bin is in a row second from the top.

17. The method of claim 12 , wherein generating the reduced context bins for the bins is performed using a mask.

Assignees

• Sze Vivienne
• Budagavi Madhukar
• Texas Instruments Inc

Inventors

• Sze Vivienne
• Budagavi Madhukar

Classifications

• H04N19/134

  characterised by the element, parameter or criterion affecting or controlling the adaptive coding · CPC title

• H04N19/176

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

• H04N19/13

  Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC] · CPC title

• H04N19/436

  using parallelised computational arrangements · CPC title

• H04N19/18 Primary

  the unit being a set of transform coefficients · CPC title