Method and device for determining transform block size

1. A method for determining a transform block size in an image device, the method comprising:

obtaining by the image device, according to image block information and a splitting manner of an image block, a parameter of a transform block corresponding to the image block, wherein the image device comprises a processor; and

obtaining, by the image device, a transform block size according to the parameter of the transform block and the splitting manner of the image block;

wherein the obtaining a transform block size according to the parameter of the transform block and the splitting manner of the image block comprises that:

a horizontal size of the transform block is greater than a vertical size of the transform block when a horizontal size of a prediction block is greater than a vertical size of the prediction block; or

a vertical size of the transform block is greater than a horizontal size of the transform block when a vertical size of a prediction block is greater than a horizontal size of the prediction block; or

a horizontal size of the transform block is equal to a vertical size of the transform block when a horizontal size of a prediction block is equal to a vertical size of the prediction block.

2. The method according to claim 1 , wherein the parameter of the transform block comprises a layer number corresponding to the transform block or an indication value of the transform block size.

3. The method according to claim 2 , wherein:

the indication value of the transform block size is equal to an indication value of a size of the image block minus the layer number corresponding to the transform block; or

the indication value of the transform block size is equal to an indication value of a size of the image block minus m1, and m1 is a positive integer; or

the indication value of the transform block size is equal to an indication value of a size of the image block minus the layer number corresponding to the transform block and then minus m2, and m2 is a positive integer.

4. The method according to claim 1 , wherein:

the horizontal size of the transform block is four times the vertical size of the transform block when a type of the prediction block is 2N×N; or

the vertical size of the transform block is four times the horizontal size of the transform block when a type of the prediction block is N×2N; or

the vertical size of the transform block is equal to the horizontal size of the transform block when a type of the prediction block is N×N or 2N×2N; or

the horizontal size of the transform block is four times the vertical size of the transform block when a type of the prediction block is 2N×nU or 2N×nD; or

the vertical size of the transform block is four times the horizontal size of the transform block when a type of the prediction block is nL×2N or nR×2N.

5. The method according to claim 1 , wherein:

the horizontal size of the transform block is eight times the vertical size of the transform block when a type of the prediction block is 2N×nU or 2N×nD; or

the vertical size of the transform block is eight times the horizontal size of the transform block when a type of the prediction block is nL×2N or nR×2N.

6. A method for determining a transform block size in an image device, the method comprising:

obtaining by the image device, according to image block information and a splitting manner of an image block, a parameter of a transform block corresponding to the image block, wherein the image device comprises a processor; and

obtaining, by the image device, a transform block size according to the parameter of the transform block and the splitting manner of the image block;

wherein the obtaining a transform block size according to the parameter of the transform block and the splitting manner of the image block comprises:

obtaining a first size of the transform block according to the parameter; and

obtaining the transform block size according to the first size;

wherein a horizontal size of the transform block is equal to h times the first size, and a vertical size of the transform block is equal to v times the first size; and h and v are greater than 0.

7. The method according to claim 6 , wherein:

h is greater than v when a horizontal size of a prediction block is greater than a vertical size of the prediction block; or

h is smaller than v when a horizontal size of a prediction block is smaller than a vertical size of the prediction block; or

h is equal to v when a horizontal size of a prediction block is equal to a vertical size of the prediction block.

8. The method according to claim 7 , wherein:

h=2 and v=0.5 when a type of the prediction block is 2N×N; or

h=0.5 and v=2 when a type of the prediction block is N×2N; or

h=1 and v=1 when a type of the prediction block is N×N; or

h=2 and v=2 when a type of the prediction block is 2N×2N; or

h=2 and v=0.5 when a type of the prediction block is 2N×nU or 2N×nD; or

h=0.5 and v=2 when a type of the prediction block is nL×2N or nR×2N;

wherein N is a positive integer, and n is greater than 0 and smaller than N.

9. The method according to claim 7 , wherein:

h=2 and v=0.25 when a type of the prediction block is 2N×nU or 2N×nD; or

h=0.25 and v=2 when a type of the prediction block is nL×2N or nR×2N;

wherein N is a positive integer, and n is greater than 0 and smaller than N.

10. The method according to claim 1 , wherein, before the obtaining, according to image block information and a splitting manner of an image block, a parameter of a transform block corresponding to the image block, the method further comprises:

obtaining a splitting flag value according to a type of a prediction block and a hierarchy flag value; and the hierarchy flag value is used to indicate whether a layer-by-layer identification manner is used for the transform block; and

determining, based on the splitting flag value, whether to determine the transform block size according to the splitting manner of the image block.

11. A device for determining a transform block size in an image system, comprising:

a parameter acquiring unit, configured to obtain, according to image block information and a splitting manner of an image block, a parameter of a transform block corresponding to the image block; and

a size acquiring unit, configured to obtain a transform block size according to the parameter of the transform block and the splitting manner of the image block;

wherein a horizontal size of the transform block obtained by the size acquiring unit is greater than a vertical size of the transform block when a horizontal size of a prediction block is greater than a vertical size of the prediction block; or

an obtained vertical size of the transform block is greater than a horizontal size of the transform block when a vertical size of a prediction block is greater than a horizontal size of the prediction block; or

an obtained horizontal size of the transform block is equal to a vertical size of the transform block when a horizontal size of a prediction block is equal to a vertical size of the prediction block.

12. The device according to claim 11 , wherein the parameter of the transform block obtained by the parameter acquiring unit comprises a layer number corresponding to the transform block or an indication value of the transform block size.

13. A method for determining a transform block size in an image device, the method comprising:

obtaining, by the image device, a splitting parameter according to a splitting manner of an image block and a parameter of a transform block, wherein the image device comprises a processor; and

obtaining, by the image device, a transform block size according to the splitting parameter;

wherein

when the splitting manner of the image block is 2N×N or N×2N and a splitting layer number corresponding to the transform block is 0, a value of the splitting parameter is 1; 2N×N indicates that a current image block is split into one upper image sub-block and one lower image sub-block of a same size, and N×2N indicates that the current image block is split into one left image sub-block and one right image sub-block of a same size; or

when the splitting manner of the image block is 2N×nU, 2N×nD, nL×2N, or nR×2N and a splitting layer number corresponding to the transform block is 0 or 1, a value of the splitting parameter is 1;

wherein 2N×nU and 2N×nD indicate that a current image block is split into one upper rectangular image sub-block and one lower rectangular image sub-block of different sizes; and

nL×2N and nR×2N indicate that the current image block is split into one left rectangular image sub-block and one right rectangular image sub-block of different sizes.

14. The method according to claim 13 , wherein the parameter of the transform block comprises a splitting layer number corresponding to the transform block and an indication value of the transform block size.

15. The method according to claim 13 , wherein a maximum number of transform block layers corresponding to the transform block is obtained according to the splitting parameter.

16. A method for determining a transform block size in an image device, the method comprising:

obtaining, by the image device, a splitting parameter according to a splitting manner of an image block and a parameter of a transform block, wherein the image device comprises a processor; and

obtaining, by the image device, a transform block size according to the splitting parameter;

wherein the obtaining a splitting parameter according to a splitting manner of an image block and a parameter of a transform block, and obtaining a transform block size according to the splitting parameter further comprises:

obtaining, when the splitting manner of the image block is 2N×nU, 2N×nD, nL×2N, or nR×2N and a value of the splitting parameter is 1, a parameter identifying whether further splitting is required; and

obtaining, according to the splitting parameter and the parameter identifying whether further splitting is required, a maximum number of transform block layers corresponding to the transform block.

17. A method for determining a transform block size in an image device, the method comprising:

obtaining, by the image device, a splitting parameter according to a splitting manner of an image block and a parameter of a transform block, wherein the image device comprises a processor; and

obtaining, by the image device, a transform block size according to the splitting parameter;

wherein

when the splitting manner of the image block is 2N×N or N×2N, a maximum number of transform block layers corresponding to the transform block is 1; 2N×N indicates that a current image block is split into one upper image sub-block and one lower image sub-block of a same size, and N×2N indicates that the current image block is split into one left image sub-block and one right image sub-block of a same size; or

when the splitting manner of the image block is 2N×nU, 2N×nD, nL×2N, or nR×2N, a maximum number of transform block layers corresponding to the transform block is 2;

wherein 2N×nU and 2R×nD indicate that a current image block is split into one upper rectangular image sub-block and one lower rectangular image sub-block of different sizes; and

nL×2N and nR×2N indicate that the current image block is split into one left rectangular image sub-block and one right rectangular image sub-block of different sizes.

18. The method according to claim 16 , wherein a splitting layer number corresponding to the transform block increases by 1 when a splitting flag of the transform block is 1.

19. The method according to claim 15 , wherein a coding end does not code a splitting flag of the transform block when a splitting layer number is greater than or equal to the maximum number of transform block layers.

20. The method according to claim 15 , wherein a decoding end does not decode a splitting flag of the transform block when a splitting layer number is greater than or equal to the maximum number of transform block layers.

21. The method according to claim 16 , wherein an indication value of the size of the transform block decreases by 1 when a splitting flag of the transform block is 1.

22. The method according to claim 13 , wherein a coding end does not code a splitting flag of the transform block when the splitting parameter is 1.

23. The method according to claim 13 , wherein a decoding end does not decode a splitting flag of the transform block when the splitting parameter is 1.

24. The method according to claim 15 , wherein an indication value of the size of the transform block decreases by 1 when a splitting layer number is smaller than the maximum number of transform block layers or when a splitting flag of the transform block is 1.