JP-2026076333-A - Image decoding method, decoder, and storage medium

JP2026076333AJP 2026076333 AJP2026076333 AJP 2026076333AJP-2026076333-A

Abstract

[Problem] To provide an image decoding method, decoder, and storage medium that avoid excessively fine block division, effectively reduce the amount of header information, and improve coding efficiency. [Solution] The image decoding method receives bitstream data, analyzes it to obtain coding tree units, then performs analysis on the nodes of each layer corresponding to the coding tree units to obtain state parameters and mark parameters for the layer nodes, then performs detection processing on the nodes according to the state parameters and mark parameters, and then performs decoding processing on the nodes with data to obtain all coding units corresponding to the coding tree units and obtain the corresponding decoded images. [Selection Diagram] Figure 6

Inventors

マー、イェンチュオ
フオ、チュンイェン
カオ、チアン
ヤン、フーチョン
ワン、シューアイ

Assignees

オッポ広東移動通信有限公司

Dates

Publication Date: 20260511
Application Date: 20260217

Claims (12)

An image decoding method, The process involves receiving bitstream data, analyzing the bitstream data, and obtaining a coding tree unit corresponding to the bitstream data. This involves initiating a preset refresh mode, which is used to perform refresh decoding between coding units. Analyzing a node in a layer corresponding to a coding tree unit to determine a state parameter and a mark parameter corresponding to the node, wherein the state parameter is used to determine whether the node supports refresh, the mark parameter is used to determine whether the node enables refresh, and enabling refresh means allowing refresh decoding to be performed on the pixel region corresponding to the node based on the state parameter and the mark parameter. The process includes determining the leaf parameters corresponding to the node, wherein the leaf parameters are used to determine whether or not to divide the node. Image decoding methods.
After determining the leaf parameters corresponding to the node, the image decoding method further: Based on the leaf parameters, a detection process is performed on the node to obtain the node of the next layer corresponding to the coding tree unit. The detection process is continued for the nodes of the next layer, and all coding unit data corresponding to the coding tree unit is obtained. The process includes generating a decoded image corresponding to the bitstream data based on the node and all the coding unit data, The image decoding method according to claim 1.
Based on the node and all the coding unit data, generating a decoded image corresponding to the bitstream data is: Based on the nodes, a decoding process is performed on all the coding unit data to obtain all the pixel data corresponding to the coding tree units. The process includes generating the decoded image corresponding to the bitstream data according to all the aforementioned pixel data, The image decoding method according to claim 2.
Before continuing to perform the detection process on the nodes of the next layer, the image decoding method: The further includes transmitting the state parameters of the next layer according to the leaf parameters, state parameters, and mark parameters. The image decoding method according to claim 2.
Transmitting the state parameters of the next layer according to the leaf parameters, state parameters, and mark parameters is: The process involves determining whether or not to perform a state refresh according to the leaf parameter, the state parameter, and the mark parameter, When it is determined that a default state refresh should be performed, a refresh process is performed on the state parameters to obtain the state parameters of the next layer, If it is determined that a state refresh should not be performed, the state parameter is determined to be the state parameter of the next layer, The image decoding method according to claim 4.
A decoder comprising a receiving unit, an analysis unit, a start unit, a detection unit, an acquisition unit, and a generation unit, The receiving unit is configured to receive bitstream data, The analysis unit is configured to analyze the bitstream data and obtain coding tree units corresponding to the bitstream data. The start unit is configured to initiate a preset refresh mode, which is used to perform refresh decoding between coding units. The analysis unit is further configured to analyze a node in a layer corresponding to a coding tree unit to determine a state parameter and a mark parameter corresponding to the node, wherein the state parameter is used to determine whether the node supports refresh, and the mark parameter is used to determine whether the node enables refresh, and enabling refresh means allowing refresh decoding to be performed on the pixel region corresponding to the node based on the state parameter and the mark parameter. The acquisition unit is configured to determine the leaf parameters corresponding to the node, and the leaf parameters are used to determine whether or not to divide the node. decoder.
The acquisition unit is configured to determine the leaf parameters corresponding to the node, then perform a detection process on the node based on the leaf parameters, and acquire the node of the next layer corresponding to the coding tree unit. The detection unit is configured to continue executing the detection process on the nodes of the next layer and to acquire all coding unit data corresponding to the coding tree unit. The generation unit is configured to generate a decoded image corresponding to the bitstream data based on the node and all the coding unit data. The decoder according to claim 6.
The generation unit is configured to perform a decoding process on all the coding unit data based on all the nodes to obtain all the pixel data corresponding to the coding tree unit, and to generate the decoded image corresponding to the bitstream data according to all the pixel data. The decoder according to claim 7.
The decoder further comprises a transmission unit, The transmission unit is configured to transmit the state parameters of the next layer according to the leaf parameters, state parameters, and mark parameters before continuing to perform the detection process for the nodes of the next layer. The decoder according to claim 7.
The transmission unit determines whether or not to perform a state refresh according to the leaf parameter, the state parameter, and the mark parameter, When it is determined that a default state refresh should be performed, a refresh process is performed on the state parameters to obtain the state parameters of the next layer, If it is determined that a state refresh should not be performed, the system is configured to determine the state parameter to be the state parameter of the next layer, The decoder according to claim 9.
A decoder comprising a processor, a memory for storing executable instructions of the processor, a communication interface, and a bus connecting the processor, the memory, and the communication interface, The decoder, which, when the instruction is executed by the processor, implements the image decoding method according to any one of claims 1 to 5.
A computer-readable storage medium for storing a program and a bitstream, wherein the program causes a processor to execute the image decoding method described in any one of claims 1 to 5, thereby decoding the bitstream and generating a decoded image.

Description

This embodiment relates to the video codec technology, and more particularly to an image decoding method, a decoder, and a storage medium. Currently, the technical solution for video coding primarily involves using partitioning techniques to divide the spatial region of an image into small, non-overlapping blocks, which are then used as the basic units for coding. The multi-type tree partitioning technique (MTT) used here has evolved stepwise from the quadtree partitioning technique (QT), to the quadtree and binary tree partitioning technique (QTBT), and further to the ternary tree partitioning technique (TT). Therefore, the distinction between QT, QTBT, and MTT refers to different partitioning methods used when performing image partitioning; the underlying partitioning principle is the same for all three. To achieve better coding performance, video coding always requires subdividing each frame of the image. However, in current video coding techniques with fixed subdivision schemes, finer subdivisions generate more header and repeating information, thereby reducing coding efficiency. This is a schematic diagram 1 showing the partitioning using QT technology.This is a schematic diagram 2 showing the division using QT technology.This is a schematic diagram of excessive division.This is an illustrative structural diagram of the video coding system configuration.This is an illustrative structural diagram of the video decoding system configuration.This is an illustrative implementation flowchart of the image decoding method according to the present embodiment.This is an illustrative structural diagram of a coding tree unit in the prior art.This is an illustrative structural diagram of a coding tree unit in the present embodiment.This is flowchart 1 illustrating an example of an image decoding method according to the present embodiment.This is flowchart 2 illustrating an example of an image decoding method according to the present embodiment.This is a schematic diagram of the state parameters and mark parameters in the embodiment of the present invention.This is a schematic diagram of the partitioning process (Figure 1).This is a schematic diagram of the partitioning process (Figure 2).This is a schematic diagram of the partitioning process (Figure 3).This is a schematic diagram of the partitioning process (Figure 4).This is a schematic diagram of the partitioning process (Figure 5).This is a schematic diagram of the partitioning process (Figure 6).This is a schematic diagram of the image coding method according to the present embodiment.This is a schematic diagram 1 that does not involve overlapping divisions.This is a schematic diagram 2 that does not involve duplicate divisions.This is a schematic diagram 1 showing the overlapping division.This is a schematic diagram 2 showing the overlapping division.Figure 1 shows an exemplary structure of the decoder configuration according to the present embodiment.This is a structural diagram 2 illustrating the configuration of the decoder according to the present embodiment. The technical solutions in these embodiments will be clearly and completely described below with reference to the drawings. It should be understood that the specific embodiments described herein are used solely to illustrate the associated applications and are not intended to limit those applications. Furthermore, for ease of explanation, only the parts relevant to the associated applications are shown in the drawings. Coding a video is like coding each frame of an image, and similarly, decoding a video bitstream after it has been coded and compressed is like decoding the bitstream of each frame of an image. In almost all international standards for video image coding, when coding a single frame of an image, it is necessary to divide the image into several M x M pixel sub-images, called coding units (CUs). Each sub-image is coded one block at a time, with the CU being the basic coding unit. The usual sizes of M are 4, 8, 16, 32, and 64. Therefore, coding a video image sequence means sequentially coding each coding unit, i.e., each CU, for each frame of the image. Decoding the bitstream of a video image sequence also means sequentially decoding each CU for each frame of the image, and finally, the entire video image sequence is reconstructed. To adapt to the different content and properties of each part of an image within a single frame, targeted and most effective coding is performed, and the size of each CU within a single frame of an image may vary, some being 8x8 and others 64x64. To seamlessly connect CUs of different sizes, a single frame of an image is typically first divided into the largest coding unit (LCU) or coding tree unit (CTU) having NxN pixels of the same size, and then each LCU is further divided into multiple CUs of varying sizes. For example, a single frame of an image is first divided into LCUs of the same size, 64x64 pixels, i.e., N=64. Here, a particular LCU consists of three 32