Granted Patent B2

US 12,445,614 · App. 17/985,809 · Granted Oct 14, 2025

Inventors: Xiaozhong Xu (State College, PA); Shan Liu (San Jose, CA)

Assignee: Tencent America LLC

H04N19/13H04N19/105H04N19/172H04N19/1883H04N19/70H04N19/91

Abstract

Claims (46)

1. A method of decoding in a decoder, comprising:

receiving a video bitstream including a current picture and a previous picture, wherein the previous picture comprises multiple regions, each region of the multiple regions in the previous picture being entropy coded, the multiple regions in the previous picture including coding tree units (CTUs) located at four corners and a center of the previous picture, respectively, the current picture including four corner regions and a center region;

determining previous probability information associated with each region of the multiple regions in the previous picture, the previous probability information being determined with a second quantization precision;

storing the previous probability information with a first quantization precision that is lower than the second quantization precision, the first quantization precision having less bits than the second quantization precision;

selecting, based on a location of an independently decodable coding segment that is a region in the current picture, one of the multiple regions in the previous picture by one of:

selecting a CTU located at one of the four corners of the previous picture when a block in the independently decodable coding segment is located at one of the four corner regions of the current picture, the CTU located at the one of the four corners being in the selected one of the multiple regions, and

selecting a CTU located at the center of the previous picture when the block in the independently decodable coding segment is located at the center region in the current picture, the CTU located at the center being in the selected one of the multiple regions;

determining initial probability information for entropy decoding one or more current syntax elements of the block in the independently decodable coding segment in the current picture based on the stored previous probability information that has the first quantization precision and is associated with the selected one of the multiple regions in the previous picture, the one or more current syntax elements being first to be entropy decoded among syntax elements in the independently decodable coding segment in the current picture; and

entropy decoding coded bits associated with the one or more current syntax elements into a bin string associated with the one or more current syntax elements based on the initial probability information and a probability estimation model for the coded bits, wherein

a flag in the video bitstream indicates that the initial probability information is to be determined based on the stored previous probability information from the previous picture.

2. The method of claim 1 , wherein the independently decodable coding segment is one of an intra slice (I-slice), a bi-predictive slice (a B-slice), and a predictive slice (a P-slice) in the current picture.

3. The method of claim 1 , comprising:

obtaining one or more parameters in the probability estimation model based on a probability estimation model used to entropy decode the selected region one of the multiple regions.

4. The method of claim 1 , comprising:

obtaining adaptation rates and adaptive weights in the probability estimation model based on a probability estimation model used to entropy decode the selected one of the multiple regions, the adaptation rates and the adaptive weights being associated with multiple probability estimates of the bin string, each of the adaptation rates associated with a respective probability estimate indicating a quantization step size between two adjacent probabilities for the bin string.

5. The method of claim 1 , further comprising:

selecting the previous picture from multiple pictures in the video bitstream such that a first level in a hierarchical temporal structure associated with the previous picture is lower than a second level in the hierarchical temporal structure associated with the current picture, each picture at the first level having been decoded prior to decoding each of one or more pictures at the second level.

6. The method of claim 1 , wherein

the probability estimation model is a context-adaptive model; and

the entropy decoding includes entropy decoding the coded bits with context-adaptive binary arithmetic coding (CABAC).

7. An apparatus for encoding, comprising:

processing circuitry configured to:

determine previous probability information associated with each region of multiple regions in a previous picture, the previous probability information being determined with a second quantization precision, each region of the multiple regions in the previous picture being entropy coded, the multiple regions in the previous picture including coding tree units (CTUs) located at four corners and a center of the previous picture, respectively;

store the previous probability information with a first quantization precision that is lower than the second quantization precision, the first quantization precision having less bits than the second quantization precision;

select, based on a location of an independently decodable coding segment that is a region in a current picture including four corner regions and a center region, one of the multiple regions in the previous picture by one of:

selecting a CTU located at one of the four corners of the previous picture when a block in the independently decodable coding segment is located at one of the four corner regions of the current picture, the CTU located at the one of the four corners being in the selected one of the multiple regions, and

selecting a CTU located at the center of the previous picture when based the block in the independently decodable coding segment is located at the center region in the current picture, the CTU located at the center being in the selected region one of the multiple regions;

determine initial probability information for entropy encoding one or more current syntax elements of the block in the independently decodable coding segment in the current picture based on the stored previous probability information that has the first quantization precision and is associated with the selected one of the multiple regions in the previous picture, the one or more current syntax elements being first to be entropy encoded among syntax elements in the independently decodable coding segment in the current picture;

entropy encode a bin string associated with the one or more current syntax elements into encoded bits associated with the one or more current syntax elements based on the initial probability information and a probability estimation model for the bin string, and

signal a flag in a video bitstream indicating that the initial probability information is to be determined based on the stored previous probability information from the previous picture.

8. The apparatus of claim 7 , wherein the independently decodable coding segment is one of an intra slice (I-slice), a bi-predictive slice (a B-slice), and a predictive slice (a P-slice) in the current picture.

9. The apparatus of claim 7 , wherein the processing circuitry is configured to:

obtain one or more parameters in the probability estimation model based on a probability estimation model used for the selected one of the multiple regions in the previous picture.

10. The apparatus of claim 7 , wherein the processing circuitry is configured to:

obtain adaptation rates and adaptive weights in the probability estimation model based on a probability estimation model for the selected one of the multiple regions in the previous picture, the adaptation rates and the adaptive weights being associated with multiple probability estimates of the bin string, each of the adaptation rates associated with a respective probability estimate indicating a quantization step size between two adjacent probabilities for the bin string.

11. The apparatus of claim 7 , wherein the processing circuitry is configured to:

select the previous picture from multiple pictures such that a first level in a hierarchical temporal structure associated with the previous picture is lower than a second level in the hierarchical temporal structure associated with the current picture, each picture at the first level having been encoded prior to encoding each of one or more pictures at the second level.

12. A non-transitory computer-readable storage medium storing a video bitstream that is generated by a video encoding method, the video encoding method comprising:

determining previous probability information associated with each region of multiple regions in a previous picture, the previous probability information being determined with a second quantization precision, each region of the multiple regions in the previous picture being entropy coded, the multiple regions in the previous picture including coding tree units (CTUs) located at four corners and a center of the previous picture, respectively;

storing the previous probability information with a first quantization precision that is lower than the second quantization precision, the first quantization precision having less bits than the second quantization precision;

selecting, based on a location of an independently decodable coding segment that is a region in a current picture including four corner regions and a center region, one of the multiple regions in the previous picture by one of:

selecting a CTU located at one of the four corners of the previous picture when a block in the independently decodable coding segment is located at one of the four corner regions of the current picture, the CTU located at the one of the four corners being in the selected one of the multiple regions, and

selecting a CTU located at the center of the previous picture when the block in the independently decodable coding segment is located at the center region in the current picture, the CTU located at the center being in the selected one of the multiple regions;

determining initial probability information for entropy encoding one or more current syntax elements of the block in the independently decodable coding segment in the current picture based on the stored previous probability information that has the first quantization precision and is associated with the selected one of the multiple regions in the previous picture, the one or more current syntax elements being first to be entropy encoded among syntax elements in the independently decodable coding segment in the current picture;

entropy encoding a bin string associated with the one or more current syntax elements into encoded bits associated with the one or more current syntax elements based on the initial probability information and a probability estimation model for the bin string, and

signalling a flag in the video bitstream indicating that the initial probability information is to be determined based on the stored previous probability information from the previous picture.

Assignments (1)

Continuity (2)

References Cited (13)