Selecting a coding method for suffix values for displacement vector differences based on value intervals

Patent metadata

Field	Value
Publication number	US-12413768-B2
Application number	US-202217969633-A
Country	US
Kind code	B2
Filing date	Oct 19, 2022
Priority date	Dec 4, 2020
Publication date	Sep 9, 2025
Grant date	Sep 9, 2025

Abstract

A video decoding method includes determining, from among plural value intervals, a first value interval to which an absolute value of a displacement vector difference of a current decoding unit in a first direction belongs. The method further includes determining, based on the first value interval, that the displacement vector difference in the first direction has a prefix value and a suffix value, and determining, based on the first value interval, a first suffix decoding method for decoding the suffix value. The method further includes performing decoding processing on a first suffix part in a bitstream using the determined first suffix decoding method, to obtain the suffix value corresponding to the displacement vector difference in the first direction. The method further includes determining, based on the prefix value and the suffix value, the displacement vector difference in the first direction.

First claim

What is claimed is:

1. A video decoding method, performed by a processor, the method comprising:
determining, from among plural value intervals, a first value interval to which an absolute value of a displacement vector difference of a current decoding unit in a first direction belongs;
determining, based on the first value interval, that the displacement vector difference in the first direction has a prefix value and a suffix value;
determining, based on the first value interval, a first type of suffix decoding method from a plurality of different types of suffix decoding methods for decoding the suffix value, each of the plurality of different types of suffix decoding methods being associated with a respective one of the plural value intervals, the plurality of different types of suffix decoding methods including the first type of suffix decoding method associated with the first value interval and a second type of suffix decoding method associated with a second value interval of the plural value intervals;
performing decoding processing on a first suffix part in a bitstream using the determined first type of suffix decoding method, to obtain the suffix value corresponding to the displacement vector difference in the first direction; and
determining, based on the prefix value and the suffix value, the displacement vector difference in the first direction.

2. The method according to claim 1 , wherein the determining the first value interval comprises:
performing decoding processing on a first prefix part in the bitstream, to obtain the prefix value corresponding to the displacement vector difference in the first direction; and
determining, based on the prefix value, the first value interval among the plural value intervals.

3. The method according to claim 2 , wherein the performing decoding processing on the first prefix part in the bitstream comprises:
performing the decoding processing on a signed binary string of the first prefix part in the bitstream using a truncated unary code method, to obtain the prefix value corresponding to the displacement vector difference in the first direction.

4. The method according to claim 1 , wherein the determining the first value interval comprises:
performing decoding processing on a first prefix part in the bitstream, to obtain the prefix value corresponding to the displacement vector difference in the first direction; and
when the prefix value is in a first value range, determining the first value interval based on the prefix value; and
when the prefix value is in a second value range,
decoding the bitstream to obtain a second range identifier, wherein the second range identifier indicates whether the absolute value of the displacement vector difference in the first direction is in a second range; and
determining, based on the second range identifier, the first value interval among the plural value intervals.

5. The method according to claim 1 , wherein the determining the first value interval comprises:
decoding the bitstream to obtain a third range identifier, wherein the third range identifier indicates whether the absolute value of the displacement vector difference in the first direction is in a third range;
when the third range identifier indicates that the absolute value of the displacement vector difference in the first direction is in the third range, determining the first value interval based on the third range; and
when the third range identifier indicates that the absolute value of the displacement vector difference in the first direction is not in the third range,
performing decoding processing on a first prefix part in the bitstream, to obtain the prefix value corresponding to the displacement vector difference in the first direction; and
determining, based on the prefix value, the first value interval among the plural value intervals.

6. The method according to claim 1 , wherein the determining the first type of suffix decoding method comprises:
when the first value interval meets a first condition, determining that the first type of suffix decoding method is a fixed-length code decoding method; and
when the first value interval meets a second condition, determining that the first type of suffix decoding method is an exponential-Golomb decoding method.

7. The method according to claim 1 , wherein
the current decoding unit corresponds to an intra block copy (IBC) decoding block, and the displacement vector difference is a block vector difference (BVD); or
the current decoding unit corresponds to an intra string copy (ISC) decoding block, and the displacement vector difference is a string vector difference (SVD); or
the current decoding unit corresponds to a decoding block in an inter-frame prediction mode, and the displacement vector difference is a motion vector difference (MVD).

8. The method according to claim 1 , wherein the plural value intervals are set based on a distribution characteristic of the displacement vector difference.

9. The method according to claim 1 , further comprising:
determining, from the plural value intervals, the second value interval, to which an absolute value of a displacement vector difference of the current decoding unit in a second direction belongs;
determining, based on the second value interval, that the displacement vector difference in the second direction has a prefix value and a suffix value;
determining, based on the second value interval, the second type of suffix decoding method for decoding the suffix value of the displacement vector difference in the second direction;
performing decoding processing on a second suffix part in the bitstream using the determined second type of suffix decoding method, to obtain the suffix value corresponding to the displacement vector difference in the second direction; and
determining, based on the prefix value and the suffix value corresponding to the displacement vector difference in the second direction, the displacement vector difference in the second direction, wherein
(i) the first direction is a horizontal direction and the second direction is a vertical direction; or (ii) the first direction is the vertical direction and the second direction is the horizontal direction.

10. The method according to claim 9 , wherein
the prefix value corresponding to the displacement vector difference in the second direction is entropy coded with context-based adaptive binary arithmetic coding (CABAC), and a first bit of the prefix value corresponding to the displacement vector difference in the second direction has a first context model and a second context model; and
when the displacement vector difference in the first direction is decoded first and the displacement vector difference in the second direction is decoded second, the method further comprises:
when the displacement vector difference in the first direction is 0, determining a value of the first bit of the prefix value corresponding to the displacement vector difference in the second direction according to the first context model; and
when the displacement vector difference in the first direction is not 0, determining the value of the first bit of the prefix value corresponding to the displacement vector difference in the second direction according to the second context model.

11. The method according to claim 1 , wherein the determining, based on the prefix value and the suffix value, the displacement vector difference in the first direction comprises:
determining, based on the prefix value corresponding to the displacement vector difference in the first direction, a first offset value;
determining, based on the first offset value and the suffix value corresponding to

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Hu Ye

Classifications

• H04N19/46

  Embedding additional information in the video signal during the compression process (H04N19/517, H04N19/68, H04N19/70 take precedence) · CPC title

• H04N19/44

  Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder · CPC title

• H04N19/184

  the unit being bits, e.g. of the compressed video stream · CPC title

• H04N19/513 Primary

  Processing of motion vectors · CPC title

• H04N19/91

  Entropy coding, e.g. variable length coding [VLC] or arithmetic coding · CPC title