Granted Patent B2

US 12,341,987 · App. 18/401,806 · Granted Jun 24, 2025

Inventors: Junyan Huo (Guangdong, CN); Yanzhuo Ma (Guangdong, CN); Shuai Wan (Guangdong, CN); Fuzheng Yang (Guangdong, CN); Qihong Ran (Guangdong, CN)

Assignee: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP., LTD.

H04N19/44H04N19/105H04N19/132H04N19/159H04N19/176

Abstract

Claims (75)

1. A prediction decoding method, comprising:

determining first multiple neighbouring reference samples of a coding block, wherein the first multiple neighbouring reference samples comprise reference samples in a reference row or a reference column neighbouring the coding block;

for the reference row, determining a value of K according to a length of a top side of the coding block and a first preset proportion, or for the reference column, determining the value of K according to a length of a left side of the coding block and a second preset proportion;

determining, from a starting position of the reference row or the reference column, positions corresponding to K reference samples, wherein K is a positive integer;

determining reference sample positions for the second multiple neighbouring reference samples based on sampling intervals, wherein a sampling starting position corresponding to the sampling intervals is indicated by the value of K;

determining second multiple neighbouring reference samples according to the determined positions corresponding to the K reference samples, wherein the second multiple neighbouring reference samples comprise neighbouring reference samples among the first multiple neighbouring reference samples;

performing prediction decoding on the coding block according to the second multiple neighbouring reference samples;

determining reference sample positions according to a preset number of samples, based on the second multiple neighbouring reference samples, wherein reference sample positions for the second multiple neighbouring reference samples are determined based on a function including a start position and a variable representing the preset number of samples; and

determining, according to the reference sample positions, reference samples corresponding to the reference sample positions from the second multiple neighbouring reference samples, to obtain third multiple neighbouring reference samples comprising the determined reference samples.

2. The method of claim 1 , wherein a prediction mode of the coding block is a multi-directional linear model (MDLM) prediction mode, wherein the MDLM prediction mode comprises an MDLM_A mode and an MDLM_L mode.

3. The method of claim 1 , wherein determining, from the starting position of the reference row or the reference column, the positions corresponding to the K reference samples comprises:

for the reference row, determining, from a leftmost position of the reference row to the right, positions corresponding to K consecutive reference samples; or

for the reference column, determining, down from a topmost position of the reference column, positions corresponding to K consecutive reference samples.

4. The method of claim 1 , wherein performing prediction decoding on the coding block according to the second multiple neighbouring reference samples comprises:

determining a second model parameter according to the third multiple neighbouring reference samples;

establishing a second prediction model according to the second model parameter, wherein the second prediction model is representative of a prediction relationship between a first colour component and a second colour component corresponding to the coding block; and

performing prediction decoding on the coding block according to the second prediction model.

5. The method of claim 1 , wherein determining reference sample positions according to a preset number of samples comprises:

determining the reference sample positions in second multiple neighbouring reference samples according to sampling intervals.

6. The method of claim 5 , wherein the sampling intervals are equal sampling intervals, determining the reference sample positions in second multiple neighbouring reference samples according to sampling intervals comprises:

determining the reference sample positions in second multiple neighbouring reference samples by performing uniform sampling on the second multiple neighbouring reference samples according to the equal sampling intervals and the following function:

(startPosN+pos*pickStepN)

wherein startPosN represents the start position of the reference sample positions, pos represents the determining procedure from position 0 to position cntN-1 with variable entN represents the preset number of samples, and pickStepN represents the equal sampling interval.

7. The method of claim 5 , wherein the sampling intervals are unequal sampling intervals, determining the reference sample positions in second multiple neighbouring reference samples according to sampling intervals comprises:

determining the reference sample positions in second multiple neighbouring reference samples by performing non-uniform sampling on the second multiple neighbouring reference samples according to the unequal sampling intervals.

8. The method of claim 2 , wherein the MDLM_A mode is INTRA_T_CCLM mode, and the MDLM_L mode is INTRA_L_CCLM mode.

9. The method of claim 8 , wherein in the INTRA_T_CCLM mode, the reference row neighbouring the coding block refers to an above neighbouring row of the coding block, in the INTRA_L_CCLM mode, the reference column neighbouring the coding block refers a left neighbouring row of the coding block.

10. The method of claim 1 , wherein each reference sample corresponds to a luma position, the neighbouring reference samples comprises above neighbouring samples and left neighbouring reference samples.

11. A prediction decoding apparatus, comprising:

a processor; and

a memory configured to store computer programs which, when executed by the processor, are operable with the processor to:

determine first multiple neighbouring reference samples of a coding block, wherein the first multiple neighbouring reference samples comprise reference samples in a reference row or a reference column neighbouring the coding block;

for the reference row, determine a value of K according to a length of a top side of the coding block and a first preset proportion, or for the reference column, determine the value of K according to a length of a left side of the coding block and a second preset proportion;

determine, from a starting position of the reference row or the reference column, positions corresponding to K reference samples, wherein K is a positive integer;

determine reference sample positions for the second multiple neighbouring reference samples based on sampling intervals, wherein a sampling starting position corresponding to the sampling intervals is indicated by the value of K;

determine second multiple neighbouring reference samples according to the determined positions corresponding to the K reference samples, wherein the second multiple neighbouring reference samples comprise neighbouring reference samples among the first multiple neighbouring reference samples;

perform prediction decoding on the coding block according to the second multiple neighbouring reference samples;

determine reference sample positions according to a preset number of samples, based on the second multiple neighbouring reference samples, wherein reference sample positions for the second multiple neighbouring reference samples are determined based on a function including a start position and a variable representing the preset number of samples; and

determine, according to the reference sample positions, reference samples corresponding to the reference sample positions from the second multiple neighbouring reference samples, to obtain third multiple neighbouring reference samples comprising the determined reference samples.

12. The apparatus of claim 11 , wherein the processor configured to perform prediction decoding on the coding block according to the second multiple neighbouring reference samples is configured to:

determine a second model parameter according to the third multiple neighbouring reference samples;

establish a second prediction model according to the second model parameter, wherein the second prediction model is representative of a prediction relationship between a first colour component and a second colour component corresponding to the coding block; and

perform prediction decoding on the coding block according to the second prediction model.

13. The apparatus of claim 11 , wherein a prediction mode of the coding block is a multi-directional linear model (MDLM) prediction mode, wherein the MDLM prediction mode comprises an MDLM_A mode and an MDLM_L mode.

14. The apparatus of claim 11 , wherein the processor configured to determine, from the starting position of the reference row or the reference column, the positions corresponding to the K reference samples is configured to:

for the reference row, determine, from a leftmost position of the reference row to the right, positions corresponding to K consecutive reference samples; or

for the reference column, determine, down from a topmost position of the reference column, positions corresponding to K consecutive reference samples.

15. A non-transitory computer readable storage medium storing program codes which, when executed by a computer, are operable with the computer to perform the following operations:

determining first multiple neighbouring reference samples of a coding block, wherein the first multiple neighbouring reference samples comprise reference samples in a reference row or a reference column neighbouring the coding block;

for the reference row, determining a value of K according to a length of a top side of the coding block and a first preset proportion, or for the reference column, determining the value of K according to a length of a left side of the coding block and a second preset proportion;

determining, from a starting position of the reference row or the reference column, positions corresponding to K reference samples, wherein K is a positive integer;

determining reference sample positions for the second multiple neighbouring reference samples based on sampling intervals, wherein a sampling starting position corresponding to the sampling intervals is indicated by the value of K;

determining second multiple neighbouring reference samples according to the determined positions corresponding to the K reference samples, wherein the second multiple neighbouring reference samples comprise neighbouring reference samples among the first multiple neighbouring reference samples;

performing prediction decoding on the coding block according to the second multiple neighbouring reference samples;

determining reference sample positions according to a preset number of samples, based on the second multiple neighbouring reference samples, wherein reference sample positions for the second multiple neighbouring reference samples are determined based on a function including a start position and a variable representing the preset number of samples; and

determining, according to the reference sample positions, reference samples corresponding to the reference sample positions from the second multiple neighbouring reference samples, to obtain third multiple neighbouring reference samples comprising the determined reference samples.

16. The non-transitory computer readable storage medium of claim 15 ,

wherein in terms of performing prediction decoding on the coding block according to the second multiple neighbouring reference samples, the program codes are operable with the computer to perform the following:

determining a second model parameter according to the third multiple neighbouring reference samples;

establishing a second prediction model according to the second model parameter, wherein the second prediction model is representative of a prediction relationship between a first colour component and a second colour component corresponding to the coding block; and

performing prediction decoding on the coding block according to the second prediction model.

17. The non-transitory computer readable storage medium of claim 15 ,

wherein in terms of determining, from the starting position of the reference row or the reference column, the positions corresponding to the K reference samples, the program codes are operable with the computer to perform the following:

for the reference row, determining, from a leftmost position of the reference row to the right, positions corresponding to K consecutive reference samples; or

for the reference column, determining, down from a topmost position of the reference column, positions corresponding to K consecutive reference samples.

18. The non-transitory computer readable storage medium of claim 15 , wherein in terms of performing prediction decoding on the coding block according to the second multiple neighbouring reference samples, the program codes are operable with the computer to perform the following:

determining a second model parameter according to the third multiple neighbouring reference samples;

establishing a second prediction model according to the second model parameter, wherein the second prediction model is representative of a prediction relationship between a first colour component and a second colour component corresponding to the coding block; and

performing prediction decoding on the coding block according to the second prediction model.

19. The non-transitory computer readable storage medium of claim 15 , wherein in terms of determining reference sample positions according to a preset number of samples, the program codes are operable with the computer to perform the following:

determining the reference sample positions in second multiple neighbouring reference samples according to sampling intervals.

20. The non-transitory computer readable storage medium of claim 19 , wherein the sampling intervals are equal sampling intervals, in terms of determining the reference sample positions in second multiple neighbouring reference samples according to sampling intervals, the program codes are operable with the computer to perform the following:

determining the reference sample positions in second multiple neighbouring reference samples by performing uniform sampling on the second multiple neighbouring reference samples according to the equal sampling intervals and the following function:

(startPosN+pos*pickStepN)

wherein startPosN represents the start position of the reference sample positions, pos represents the determining procedure from position 0 to position cntN-1 with variable cntN represents the preset number of samples, and pickStepN represents the equal sampling interval.

Assignments (1)

Continuity (4)

References Cited (94)