VIDEO CODING METHOD, COMPUTER DEVICE, AND STORAGE MEDIUM

• US 20190320175A1
• Filed: 06/24/2019
• Published: 10/17/2019
• Est. Priority Date: 11/21/2017
• Status: Active Grant

• Current Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Original Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Inventors: MAO, Xunan
• Granted Patent: US 10,944,970 B2
• Time in Patent Office: Days
• CPC Class Codes: H04N 19/114 Adapting the group of pictures [GOP] structure, e.g. number of B-frames between two anchor frames H04N19/107 takes precedence
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/14 Coding unit complexity, e.g. amount of activity or edge presence estimation H04N19/146 takes precedence
• CPC Class Codes: H04N 19/146 Data rate or code amount at the encoder output
• CPC Class Codes: H04N 19/159 Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
• CPC Class Codes: H04N 19/172 the region being a picture, frame or field
• CPC Class Codes: H04N 19/187 the unit being a scalable video layer
• CPC Class Codes: H04N 19/30 using hierarchical techniques, e.g. scalability H04N19/63 takes precedence
• CPC Class Codes: H04N 19/31 in the temporal domain
• CPC Class Codes: H04N 19/463 by compressing encoding parameters before transmission
• CPC Class Codes: H04N 19/573 Motion compensation with multiple frame prediction using two or more reference frames in a given prediction direction
• CPC Class Codes: H04N 19/577 Motion compensation with bidirectional frame interpolation, i.e. using B-pictures

Abstract

A video coding method can include receiving, by processing circuitry of a coding device, a current picture in a group of pictures (GOP) of a layered coding structure in video sequence. The current picture can have two reference pictures. A current layer of the current picture in the layered coding structure can be determined. A current quantization parameter (QP) offset of the current picture can be determined according to the current layer of the current picture. Reference QPs corresponding to the two reference pictures of the current picture can be determined. A current QP of the current picture can be determined according to the reference QPs and the current QP offset. The current picture can be coded according to the current QP.

Claims

1. A video coding method, comprising:

receiving, by processing circuitry of a coding device, a current picture in a group of pictures (GOP) of a layered coding structure in video sequence, the current picture having two reference pictures;

determining, by the processing circuitry, a current layer of the current picture in the layered coding structure;

determining, by the processing circuitry, a current quantization parameter (QP) offset of the current picture according to the current layer of the current picture;

determining, by the processing circuitry, reference QPs corresponding to the two reference pictures of the current picture;

determining, by the processing circuitry, a current QP of the current picture according to the reference QPs and the current QP offset; and

coding the current picture according to the current QP.

2. The method according to claim 1, wherein the two reference pictures are B frames, and the determining the current QP of the current picture comprises:

determining a first distance between a first reference picture of the two reference pictures and the current picture, and a second distance between a second reference picture of the two reference pictures and the current picture;

determining a first QP based on the first and second distances and the reference QPs of the two reference pictures; and

determining the current QP of the current picture based on a sum of the first QP and the current QP offset.

3. The method according to claim 2, wherein the determining the first QP comprises:

determining the first QP based on a weighted sum of the reference QPs of the two reference pictures based on a first weight and a second weight,

wherein the first weight corresponds to the first reference picture, and is a first ratio of the second distance to a sum of the first and second distances, and

the second weight corresponds to the second reference picture, and is a second ratio of the first distance to the sum of the first and second distances.

4. The method according to claim 1, wherein the two reference pictures are I frames, and the determining the current QP of the current picture comprises:

determining an average of the reference QPs of the two reference pictures;

determining a second QP that is a sum of the average and a predetermined value; and

determining the current QP of the current picture based on a sum of the second QP and the current QP offset.

5. The method according to claim 1, wherein a first reference picture of the two reference pictures is an I frame, and a second reference picture of the two reference pictures is a P frame or B frame, and the determining the current QP of the current picture comprises:

determining a third QP that is the QP of the second reference picture; and

determining the current QP of the current picture to be a sum of the third QP and the current QP offset.

6. The method according to claim 1, wherein the determining the reference QPs comprises:

determining a type and a QP of a first reference picture of the two reference pictures, the type being one of I frame, P frame, and B frame; and

determining the reference QP of the first reference picture according to the type and the QP of the first reference picture, the reference QP of the first reference picture being one of the reference QPs.

7. The method according to claim 6, wherein the determining the reference QP of the first reference picture comprises:

when the type of the first reference picture is B frame,

determining a layer of the first reference picture in the layer coding structure;

determining a reference QP offset corresponding to the first reference picture according to the layer of the first reference picture; and

determining the reference QP of the first reference picture based on a difference between the QP of the first reference picture and the reference QP offset of the first reference picture.

8. The method according to claim 6, wherein the determining the reference QP of the first reference picture comprises:

when the type of the first reference picture is I frame or P frame,

determining a QP of the first reference picture to be the first reference QP.

9. The method according to claim 1, wherein the coding the current picture comprises:

determining a current quantization step corresponding to the current picture according to the current QP; and

coding the current picture frame according to the current quantization step to obtain a current bit rate corresponding to the current picture.

10. The method according to claim 9, further comprising:

determining a quantization offset parameter of the current layer of the current picture with respect to a unidirectionally predicted picture in the GOP;

determining a complexity update value according to the current bit rate of the current picture and the quantization offset parameter; and

determining an updated picture complexity parameter according to a picture complexity parameter of a preceding picture frame of the current picture frame and the picture complexity update value.

11. The method according to claim 10, further comprising:

determining a type of a subsequent picture that is coded after the current picture; and

when the subsequent picture is a non-bidirectionally predicted picture, determining a QP corresponding to the subsequent picture according to a bit rate control model and the updated picture complexity parameter.

12. The method according to claim 1, wherein the determining the current QP offset comprises:

determining a quantization offset parameter of the current layer with respect to a unidirectionally predicted picture in the GOP; and

determining the current QP offset according to the quantization offset parameter.

13. The method according to claim 12, wherein a first layer in the layered coding structure that is higher than a second layer in the layered coding structure has a higher quantization offset parameter.

14. An apparatus of video coding, comprising circuitry configured to:

receive a current picture in a group of pictures (GOP) of a layered coding structure in video sequence, the current picture having two reference pictures;

determine a current layer of the current picture in the layered coding structure;

determine a current quantization parameter (QP) offset of the current picture according to the current layer of the current picture;

determine reference QPs corresponding to the two reference pictures of the current picture;

determine a current QP of the current picture according to the reference QPs and the current QP offset; and

coding the current picture according to the current QP.

18. The non-transitory computer-readable medium according to claim 14, wherein the two reference pictures are I frames, and the determining the current QP of the current picture comprises:

determining an average of the reference QPs of the two reference pictures;

determining a second QP that is a sum of the average and a predetermined value; and

determining the current QP of the current picture based on a sum of the second QP and the current QP offset.

19. The non-transitory computer-readable medium according to claim 14, wherein a first reference picture of the two reference pictures is an I frame, and a second reference picture of the two reference pictures is a P frame or B frame, and the determining the current QP of the current picture comprises:

determining a third QP that is the QP of the second reference picture; and

determining the current QP of the current picture to be a sum of the third QP and the current QP offset.

20. The non-transitory computer-readable medium according to claim 14, wherein the determining the reference QPs comprises:

determining a type and a QP of a first reference picture of the two reference pictures, the type being one of I frame, P frame, and B frame; and

determining the reference QP of the first reference picture according to the type and the QP of the first reference picture, the reference QP of the first reference picture being one of the reference QPs.

15. A non-transitory computer-readable medium storing a program executable by a processor to perform a method, the method comprising:

receiving a current picture in a group of pictures (GOP) of a layered coding structure in video sequence, the current picture having two reference pictures;

determining a current layer of the current picture in the layered coding structure;

determining a current quantization parameter (QP) offset of the current picture according to the current layer of the current picture;

determining reference QPs corresponding to the two reference pictures of the current picture;

determining a current QP of the current picture according to the reference QPs and the current QP offset; and

coding the current picture according to the current QP.

16. The non-transitory computer-readable medium according to claim 15, wherein the method further comprises:

determining a first distance between a first reference picture of the two reference pictures and the current picture, and a second distance between a second reference picture of the two reference pictures and the current picture;

determining a first QP based on the first and second distances and the reference QPs of the two reference pictures; and

determining the current QP of the current picture based on a sum of the first QP and the current QP offset.

17. The non-transitory computer-readable medium according to claim 16, wherein the determining the first QP comprises:

determining the first QP based on a weighted sum of the reference QPs of the two reference pictures based on a first weight and a second weight,

wherein the first weight corresponds to the first reference picture, and is a first ratio of the second distance to a sum of the first and second distances, and

the second weight corresponds to the second reference picture, and is a second ratio of the first distance to the sum of the first and second distances.