Reference frame encoding method and apparatus, and reference frame decoding method and apparatus

Patent metadata

Field	Value
Publication number	US-10771813-B2
Application number	US-201815965162-A
Country	US
Kind code	B2
Filing date	Apr 27, 2018
Priority date	Oct 31, 2015
Publication date	Sep 8, 2020
Grant date	Sep 8, 2020

Abstract

A reference frame decoding method includes obtaining a first reference frame, where the first reference frame is a first picture frame on which decoding reconstruction has been performed or a first interpolated picture frame obtained by pixel interpolation on the first picture frame, parsing a bitstream to obtain mapping parameters, determining to-be-determined coefficients of a preset mapping function according to the mapping parameters, obtaining, in the first reference frame according to the mapping function whose to-be-determined coefficients have been determined, a first pixel unit having a mapping relationship with a second pixel unit of a second reference frame, and assigning a pixel value of the first pixel unit to the second pixel unit.

First claim

What is claimed is:

1. A reference frame decoding method, comprising:
obtaining, by a video decoding device, a first reference frame comprising a first picture frame on which decoding reconstruction has been performed or a first interpolated picture frame obtained by pixel interpolation on the first picture frame;
parsing, by the video decoding device, a bitstream to obtain mapping parameters;
determining, by the video decoding device, to-be-determined coefficients of a preset mapping function according to the mapping parameters;
obtaining, by the video decoding device in the first reference frame according to the preset mapping function, a first pixel unit having a mapping relationship with a second pixel unit of a second reference frame;
assigning, by the video decoding device, a pixel value of the first pixel unit to the second pixel unit to construct the second reference frame;
obtaining, by the video decoding device, a reference frame list comprising a candidate reference frame of the first picture frame on which the decoding reconstruction has been performed;
obtaining, by the video decoding device, regional vertexes of a preset first region in the first picture frame on which the decoding reconstruction has been performed;
obtaining, by the video decoding device in the second reference frame according to the preset mapping function, scatters having the mapping relationship with the regional vertexes of the preset first region;
coupling, by the video decoding device according to a coupling relationship between the regional vertexes of the preset first region, the scatters in the second reference frame having the mapping relationship with the regional vertexes of the preset first region;
forming, by the video decoding device, a second region using a region encircled by the coupled scatters;
calculating, by the video decoding device, a ratio of an intersection of areas of the preset first region and the second region to a union of the areas of the preset first region and the second region, wherein the intersection comprises an overlapped location region between the preset first region and the second region, and wherein the union comprises the intersection and a non-overlapped location region between the preset first region and the second region in ranges of the preset first region and the second region;
adding, by the video decoding device, the second reference frame to the reference frame list when the ratio is less than a preset value; and
adding, by the video decoding device, the first picture frame on which the decoding reconstruction has been performed to the reference frame list when the ratio is greater than or equal to the preset value.

2. The reference frame decoding method of claim 1 , wherein the preset mapping function comprises x′=m 0 x+m 1 y+m 2 , y′=m 3 x+m 4 y+m 5 , wherein, a location of the first pixel unit is x′=m 0 x+m 1 y+m 2 , y′=m 3 x+m 4 y+m 5 , wherein x, y are a horizontal coordinate and a vertical coordinate of the second pixel unit, wherein x′, y′ are a horizontal coordinate and a vertical coordinate of the first pixel unit having the mapping relationship with the second pixel unit, and wherein m 0 ,m 1 ,m 2 ,m 3 ,m 4 ,m 5 are the to-be-determined coefficients of the preset mapping function determined according to the mapping parameters.

3. The reference frame decoding method of claim 1 , further comprising:
obtaining, by the video decoding device, a third reference frame comprising a second picture frame on which the decoding reconstruction has been performed or a second interpolated picture frame obtained by the pixel interpolation on the second picture frame;
obtaining, by the video decoding device in the third reference frame according to the preset mapping function whose the to-be-determined coefficients have been determined, a third pixel unit having the mapping relationship with a fourth pixel unit of a fourth reference frame; and
assigning, by the video decoding device, a pixel value of the third pixel unit to the fourth pixel unit.

4. The reference frame decoding method of claim 3 , wherein, a location of the third pixel unit comprises x′=m 0 x+m 1 y+m 2 , y′=m 3 x+m 4 y+m 5 , wherein x, y are a horizontal coordinate and a vertical coordinate of the fourth pixel unit, wherein x′, y′ are a horizontal coordinate and a vertical coordinate of the third pixel unit having the mapping relationship with the fourth pixel unit, and wherein m 0 ,m 1 ,m 2 ,m 3 ,m 4 ,m 5 are the to-be-determined coefficients of the preset mapping function determined according to the mapping parameters.

5. The reference frame decoding method of claim 3 , further comprising:
obtaining, by the video decoding device, a reference frame list comprising a candidate reference frame of the second picture frame on which the decoding reconstruction has been performed;
performing, by the video decoding device, weighted addition on a pixel unit of the second reference frame and a pixel unit in a same location in the fourth reference frame to obtain a fifth reference frame; and
adding, by the video decoding device, the second picture frame on which the decoding reconstruction has been performed or the fifth reference frame to the reference frame list.

6. The reference frame decoding method of claim 1 , wherein obtaining the scatters comprises obtaining, by the video decoding device, locations of the scatters as

x
′

=

-

m
4

⁢
x

+

m
1

⁢
y

+

m
2

⁢

m
4

-

m
1

⁢

m
5

m
1

⁢

m
3

-

m
0

⁢

m
4

,

⁢

y
′

=

m
3

⁢
x

Assignees

• Huawei Tech Co Ltd

Inventors

• Lin Sixin
• Zhang Xiang
• Ma Siwei
• Yang Haitao

Classifications

• H04N19/51 Primary

  Motion estimation or motion compensation · CPC title

• H04N19/513

  Processing of motion vectors · CPC title

• H04N7/24

  Systems for the transmission of television signals using pulse code modulation (H04N21/00 takes precedence) · CPC title

• H04N19/537

  Motion estimation other than block-based · CPC title

• H04N19/593

  involving spatial prediction techniques · CPC title