KR-102962448-B1 - Method and device for picture encoding and decoding

KR102962448B1KR 102962448 B1KR102962448 B1KR 102962448B1KR-102962448-B1

Abstract

A decoding method is presented. First, an illumination compensation parameter is determined for the current block of the picture. The illumination compensation parameter is determined from one or more illumination compensation parameters, or from one or more reconstructed samples of at least one spatial neighbor block only if said at least one spatial neighbor block belongs to the same local illumination compensation group as the current block, which is called the current local illumination compensation group. Finally, the current block is reconstructed using the determined illumination compensation parameter.

Inventors

천, 야
푸아리에, 땅기
보르드, 필립
르레아넥, 파브리쓰
갈팡, 프랑크

Assignees

인터디지털 브이씨 홀딩스 인코포레이티드

Dates

Publication Date: 20260511
Application Date: 20190919
Priority Date: 20181210

Claims (20)

As a decoding method: A step of acquiring a current picture divided into multiple coding units - each coding unit among the multiple coding units belongs to a single local lighting compensation group among multiple local lighting groups dividing the current picture, and the local lighting compensation group is an area of the current picture composed of a subset of adjacent samples of the current picture -; A step of determining at least one lighting compensation parameter representing a local lighting compensation linear model for a current coding unit among a plurality of coding units of a local lighting group; and The method includes the step of reconstructing the current coding unit using the determined lighting compensation parameter. For any current coding unit of any current local lighting compensation group of the current picture, information of the current picture used to determine the at least one lighting compensation parameter representing the local lighting compensation linear model for the current coding unit comprises one or more reconstructed samples of at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs, or one or more lighting compensation parameters representing the local lighting compensation linear model associated with at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs.
In paragraph 1, The above local lighting compensation group is smaller than, larger than, or the same as the coding tree unit and decoding method.
In paragraph 1, A decoding method further comprising the step of decoding a flag indicating whether to reconfigure the current coding unit using the determined lighting compensation parameter in response to the fact that the current coding unit is not located in the top-left position of the current local lighting compensation group.
In paragraph 1, A decoding method further comprising the step of decoding a flag indicating whether to reconfigure the current coding unit using the determined lighting compensation parameter in response to at least one coding unit located above or to the left of the current coding unit being coded in inter mode.
In paragraph 1, A decoding method further comprising the step of decoding a flag indicating whether at least one coding unit of the current local lighting compensation group uses lighting compensation for prediction.
In paragraph 1, A decoding method further comprising the step of decoding a flag indicating whether lighting compensation parameters are calculated and stored after reconfiguring the coding units of the current local lighting compensation group.
In paragraph 1, A decoding method further comprising the step of decoding a flag indicating whether at least one coding unit of the current local lighting compensation group has a value of another flag that is true in response to another flag being explicitly coded, wherein the other flag indicates whether reconfiguring the current coding unit uses the determined lighting compensation parameter.
In paragraph 5, The above flag is a decoding method that is decoded by the first coding unit of the above current local lighting compensation group.
In paragraph 5, The above flag is a decoding method that is decoded into the first coding unit of the current local light compensation group that is not coded in merge mode.
In paragraph 1, A decoding method further comprising the step of decoding for the current local lighting group a flag indicating whether at least one coding unit in a group whose size is smaller than a threshold uses lighting compensation for computing prediction.
In paragraph 1, A decoding method further comprising the step of decoding for the current local lighting group a flag indicating whether at least one coding unit in a group whose size is greater than a threshold uses lighting compensation for computing prediction.
In paragraph 1, A decoding method in which at least one illumination compensation parameter for the current coding unit is determined from one or more illumination compensation parameters or from one or more reconstructed samples of at least one spatial neighbor coding unit that belongs to the current local illumination compensation group and has the same reference index as the current coding unit.
In paragraph 1, A decoding method in which at least one lighting compensation parameter for the current coding unit is determined as the average of lighting compensation parameters of two spatial neighbor coding units that belong to the current local lighting compensation group and have the same reference index as the current coding unit.
As a decoding device comprising one or more processors: The above one or more processors Acquire a current picture divided into multiple coding units - each coding unit among the multiple coding units belongs to a single local lighting compensation group among multiple local lighting groups dividing the current picture, and the local lighting compensation group is an area of the current picture composed of a subset of adjacent samples of the current picture -; It is configured to perform the reconstruction of the current coding unit using determined lighting compensation parameters, and A decoding device comprising, for any current coding unit of any current local lighting compensation group of the current picture, information of the current picture used to determine at least one lighting compensation parameter representing a local lighting compensation linear model for the current coding unit, one or more reconstructed samples of at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs, or one or more lighting compensation parameters representing a local lighting compensation linear model associated with at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs.
As an encoding method: A step of acquiring a current picture divided into multiple coding units - each coding unit among the multiple coding units belongs to a single local lighting compensation group among multiple local lighting groups dividing the current picture, and the local lighting compensation group is an area of the current picture composed of a subset of adjacent samples of the current picture -; A step of determining at least one lighting compensation parameter representing a local lighting compensation linear model for a current coding unit among a plurality of coding units of a local lighting group; and The method includes the step of reconstructing the current coding unit using the determined lighting compensation parameter. For any current coding unit of any current local lighting compensation group of the current picture, the information of the current picture used to determine the at least one lighting compensation parameter representing the local lighting compensation linear model for the current coding unit comprises one or more reconstructed samples of at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs, or one or more lighting compensation parameters representing the local lighting compensation linear model associated with at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs.
As an encoding device comprising one or more processors: The above one or more processors Acquire a current picture divided into multiple coding units - each coding unit among the multiple coding units belongs to a single local lighting compensation group among multiple local lighting groups dividing the current picture, and the local lighting compensation group is an area of the current picture composed of a subset of adjacent samples of the current picture -; Determining at least one lighting compensation parameter representing a local lighting compensation linear model for the current coding unit among a plurality of coding units of a local lighting group; It is configured to perform the reconstruction of the current coding unit using the lighting compensation parameter determined above, and For any current coding unit of any current local lighting compensation group of the current picture, the information of the current picture used to determine the at least one lighting compensation parameter representing the local lighting compensation linear model for the current coding unit comprises one or more reconstructed samples of at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs, or one or more lighting compensation parameters representing the local lighting compensation linear model associated with at least one reference coding unit belonging to the current local lighting compensation group to which the current coding unit belongs.
A non-transient information storage medium storing program code instructions for implementing a decoding method according to claim 1.
A non-transient information storage medium storing program code instructions for implementing the encoding method according to paragraph 15.
delete
delete

Description

Method and device for picture encoding and decoding At least one of the embodiments relates to a method and device for picture encoding and decoding in general, and in particular, to a method and device for picture encoding and decoding using local illumination compensation. To achieve high compression efficiency, image and video coding schemes typically employ prediction and transformation to leverage spatial and temporal redundancy within video content. Generally, intra- or inter-prediction is used to utilize intra- or inter-frame correlation, and the differences between the original image block and the predicted image block—often referred to as prediction errors or prediction residuals—are transformed, quantized, and entropy-coded. During encoding, the original image block is typically partitioned/split into sub-blocks using quad-tree partitioning. To reconstruct the video, the compressed data is decoded by inverse processes corresponding to prediction, transformation, quantization, and entropy coding. Figure 1 depicts a Coding Tree Unit (CTU) partitioned into smaller units. FIG. 2 illustrates the principle of deriving local illumination compensation (LIC) parameters for the current block from neighboring reconstructed samples and corresponding juxtaposed reference samples in the case of "JEM-LIC". Figure 3 illustrates the process of calculating LIC parameters in the case of "VVC-LIC". Figure 4 shows the derivation of LIC parameters in the case of "VVC-LIC" from the LIC parameters stored above on the left, and the vertical (dark gray) and horizontal (light gray) LIC parameters stored at the frame level at 4x4 resolution on the right. Figure 5 shows a CTU of size 128x128 and stored LIC parameters of size 32x2 for the case of "VVC-LIC". Figure 6 depicts the current Coding Unit (CU) and its neighboring CUs to its left and above it. Figure 7 depicts CTUs partitioned into CUs and illustrates the principle of non-availability of LIC parameters for some of them in the case of "VVC-LIC" or the principle of non-availability of neighboring reconfigured samples in the case of "JEM-LIC". FIG. 8 illustrates a flowchart of a method for decoding CU in the case of "VVC-LIC" according to one embodiment. FIG. 9 depicts CTUs partitioned into CUs and illustrates the principle of restricting LIC parameter access according to various embodiments in the case of "VVC-LIC" or "JEM-LIC". FIGS. 10 and 11 depict CTUs partitioned into CUs and illustrate the principle of LIC-groups according to various embodiments. FIGS. 12a and FIGS. 12b illustrate flowcharts of decoding methods according to various embodiments. FIG. 13 depicts the current CU and its neighboring CUs and illustrates the principle of deriving LIC parameters according to one embodiment. FIG. 14 illustrates a block diagram of a video encoder according to an embodiment. FIG. 15 illustrates a block diagram of a video decoder according to an embodiment. FIG. 16 illustrates a block diagram of an example of a system in which various aspects and embodiments are implemented. In HEVC coding, a picture is typically partitioned into square-shaped CTUs with configurable sizes of 64x64, 128x128, or 256x256. A CTU is the root of a quad-tree partitioned into four square coding units (CUs) of the same size, that is, half the size of the parent block in width and height, as depicted in Fig. 1. A quad-tree is a tree in which a parent node can be split into four child nodes, and each child node can be the parent node for another split into four child nodes. In HEVC, a coding block (CB) is partitioned into one or more prediction blocks (PBs) and forms the root of the quad-tree partitioning into transform blocks (TBs). Corresponding to the coding block, prediction block, and transformation block, the coding unit (CU) includes a tree-structured set of prediction units (PUs) and transformation units (TUs), the PU includes prediction information for all color components, and the TU includes a residual coding syntax structure for each color component. The sizes of the CB, PB, and TB of the luminance components are applied to the corresponding CU, PU, and TU. In more recent encoding systems, the CTU is the root of a coding tree that is partitioned into coding units (CUs). A coding tree is a tree in which a parent node (usually corresponding to a CU) can be split into child nodes (e.g., 2, 3, or 4 child nodes), and each child node can be a parent node for another split into child nodes. In addition to quad-tree split modes, new split modes (binary tree symmetric split modes, binary tree asymmetric split modes, and triple tree split modes) are also defined to increase the total number of possible split modes. A coding tree has a unique root node, e.g., the CTU. The leaves of a coding tree are the terminal nodes of the tree. Each node of a coding tree represents a CU that can be further split into smaller CUs, also referred to as sub-CUs or, more generally, sub-blocks. Once the partitioning of the