Video encoding / decoding method, bitstream transmission method, and recording medium storing bitstream

Abstract

A video decoding method according to the present disclosure may include the steps of: receiving a neural-network post-filter (NNPF) supplemental enhancement information (SEI) message; determining at least one neural network available as a post-processing filter based on whether the NNPF SEI message is applied to a current picture, based on at least one neural-network post-filter characteristics (NNPFC) SEI message included in the NNPF SEI message; and determining whether a target neural-network post-processing filter applicable to the current picture is activated based on at least one neural-network post-filter activation (NNPFA) SEI message included in the NNPF SEI message, wherein the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages.

Description

[Technical Field]

[0001] The present disclosure relates to a video encoding / decoding method, a method for transmitting a bitstream, and a recording medium storing the bitstream, and in particular to a method for processing a neural network postfilter. [Background technology]

[0002] In recent years, demand for high-resolution, high-quality video, such as HD (High Definition) video and UHD (Ultra High Definition) video, has been increasing in various fields. As video data becomes higher in resolution and quality, the amount of information or bits to be transmitted increases compared to existing video data. The increase in the amount of information or bits to be transmitted leads to an increase in transmission costs and storage costs.

[0003] Therefore, there is a demand for a highly efficient video compression technique for effectively transmitting, storing, and reproducing high-resolution, high-quality video information. Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present disclosure is to provide a video encoding / decoding method and apparatus with improved encoding / decoding efficiency.

[0005] Another object of the present disclosure is to provide a method for clarifying which neural network post-filter is applied among the neural network post-filters of multiple preceding NNPFC SEI messages.

[0006] Another object of the present disclosure is to provide a non-transitory computer-readable recording medium for storing a bitstream generated by the video encoding method according to the present disclosure.

[0007] Another object of the present disclosure is to provide a non-transitory computer-readable recording medium that stores a bitstream that is received and decoded by a video decoding device according to the present disclosure and is used to restore a video.

[0008] Another object of the present disclosure is to provide a method for transmitting a bitstream generated by the video encoding method according to the present disclosure.

[0009] The technical problems to be solved by the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by a person having ordinary skill in the art to which the present disclosure pertains from the following description. [Means for solving the problem]

[0010] A video decoding method according to one embodiment of the present disclosure is a video decoding method performed by a video decoding device, the video decoding method including: acquiring a neural-network post-filter (NNPF) supplemental enhancement information (SEI) message; determining at least one neural network available as a post-processing filter based on whether the NNPF SEI message is applied to a current picture, based on at least one neural-network post-filter characteristics (NNPFC) SEI message included in the NNPF SEI message; and determining whether a target neural-network post-processing filter applicable to the current picture is activated based on at least one neural-network post-filter activation (NNPFA) SEI message included in the NNPF SEI message, wherein the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages.

[0011] A video encoding method according to another aspect of the present disclosure may be a video encoding method performed by a video encoding device, the video encoding method including: encoding at least one neural network available as a post-processing filter into at least one NNPFC (neural-network post-filter characteristics) SEI (supplemental enhancement information) message; and encoding whether a target neural-network post-processing filter applicable to a current picture is activated into at least one NNPFA (neural-network post-filter activation) SEI message, wherein, in a video decoding device, based on whether the NNPF (neural-network post-filter) SEI message is applied to the current picture, the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages, and the NNPF SEI message includes the NNPFC SEI message and the NNPFA SEI message.

[0012] A computer-readable recording medium according to yet another aspect of the present disclosure can store a bitstream generated by the video encoding method or apparatus of the present disclosure.

[0013] A transmission method according to yet another aspect of the present disclosure can transmit a bitstream generated by the video encoding method or apparatus of the present disclosure.

[0014] The above briefly summarized features of the present disclosure are merely exemplary aspects of the following detailed description of the present disclosure and are not intended to limit the scope of the present disclosure. [Effects of the Invention]

[0015] According to the present disclosure, it is possible to provide a video encoding / decoding method and apparatus with improved encoding / decoding efficiency.

[0016] Furthermore, the present disclosure can clarify which neural network post-filter to apply among the neural network post-filters of the preceding NNPFC SEI messages.

[0017] Furthermore, according to the present disclosure, it is possible to provide a non-transitory computer-readable recording medium that stores a bitstream generated by the video encoding method according to the present disclosure.

[0018] Furthermore, according to the present disclosure, it is possible to provide a non-transitory computer-readable recording medium that stores a bitstream that is received and decoded by a video decoding device according to the present disclosure and used to restore a video.

[0019] Furthermore, the present disclosure can provide a method for transmitting a bitstream generated by a video encoding method.

[0020] The effects obtained by the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those having ordinary skill in the art to which the present disclosure pertains from the following description. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a schematic diagram illustrating a video coding system to which embodiments of the present disclosure can be applied; [Figure 2] 1 is a schematic diagram illustrating a video encoding device to which an embodiment of the present disclosure can be applied. [Figure 3] FIG. 1 is a schematic diagram illustrating a video decoding device to which an embodiment of the present disclosure can be applied. [Figure 4] FIG. 10 is a diagram illustrating an interleaved method for deriving a luma channel. [Figure 5]10A and 10B are diagrams illustrating various examples of persistence and revocation of NNPFA. [Figure 6] 10A and 10B are diagrams illustrating various examples of persistence and revocation of NNPFA. [Figure 7] 10A and 10B are diagrams illustrating various examples of persistence and revocation of NNPFA. [Figure 8] 10A and 10B are diagrams illustrating various examples of persistence and revocation of NNPFA. [Figure 9] 1 is a flowchart illustrating a video encoding method to which an embodiment of the present disclosure can be applied. [Figure 10] 1 is a flowchart illustrating a video decoding method to which an embodiment of the present disclosure can be applied. [Figure 11] 1 is a flowchart illustrating a video encoding method and a video decoding method to which an embodiment of the present disclosure can be applied. [Figure 12] 1 is a flowchart illustrating a video encoding method and a video decoding method to which an embodiment of the present disclosure can be applied. [Figure 13] FIG. 1 is a diagram illustrating a content streaming system to which an embodiment of the present disclosure can be applied. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement the present disclosure, but the present disclosure may be embodied in various other forms and is not limited to the embodiments described herein.

[0023] In describing the embodiments of the present disclosure, if it is determined that a specific description of a known configuration or function may obscure the gist of the present disclosure, the detailed description thereof will be omitted. In addition, in the drawings, parts that are not related to the description of the present disclosure will be omitted, and similar parts will be designated by similar reference numerals.

[0024] In this disclosure, when one component is "coupled," "bonded," or "connected" to another component, this may include not only a direct connection, but also an indirect connection where there is another component between them. Furthermore, when one component "includes" or "has" (comprises; constitutes; constructs; sets; encompasses; contains; contains) another component, this does not exclude the other component, but means that the other component may be further included, unless otherwise specified.

[0025] In this disclosure, terms such as "first" and "second" are used only to distinguish one component from another, and do not limit the order or importance of the components unless otherwise specified. Therefore, within the scope of this disclosure, a first component in one embodiment may be referred to as a second component in another embodiment, and similarly, a second component in one embodiment may be referred to as a first component in another embodiment.

[0026] In this disclosure, components that are distinguished from one another are used to clearly describe the characteristics of each component and do not necessarily mean that the components are separate. That is, multiple components may be integrated into a single hardware or software unit, or a single component may be distributed into multiple hardware or software units. Therefore, even if not specifically stated, such integrated or distributed embodiments are also included within the scope of this disclosure.

[0027] In this disclosure, the components described in various embodiments are not necessarily essential components, and some may be optional components. Therefore, an embodiment consisting of a subset of the components described in one embodiment is also within the scope of this disclosure. Furthermore, an embodiment including other components in addition to the components described in various embodiments is also within the scope of this disclosure.

[0028] This disclosure relates to video encoding and decoding, and terms used in this disclosure may have their ordinary meanings in the technical field to which this disclosure pertains unless they are newly defined in this disclosure.

[0029] In this disclosure, a "picture" generally refers to a unit representing one video image in a specific time period, a slice / tile is a coding unit constituting a part of a picture, and one picture may be composed of one or more slices / tiles. In addition, a slice / tile may include one or more coding tree units (CTUs).

[0030] In this disclosure, a "pixel" or a "pel" may refer to the smallest unit constituting one picture (or image). A "sample" may also be used as a term corresponding to a pixel. A sample may generally represent a pixel or a pixel value, and may represent only a pixel / pixel value of a luma component, or may represent only a pixel / pixel value of a chroma component.

[0031] In this disclosure, a "unit" may refer to a basic unit of video processing. A unit may include at least one of a specific region of a picture and information related to that region. A unit may also be referred to as a "sample array," "block," or "area," depending on the situation. In general, an MxN block may include a set (or array) of samples (or sample arrays) or transform coefficients consisting of M columns and N rows.

[0032] In this disclosure, a "current block" may refer to one of a "current coding block," a "current coding unit," a "block to be coded," a "block to be decoded," or a "block to be processed." When prediction is performed, a "current block" may refer to a "current predicted block" or a "block to be predicted." When transformation (inverse transformation) / quantization (inverse quantization) is performed, a "current block" may refer to a "current transformed block" or a "block to be transformed." When filtering is performed, a "current block" may refer to a "block to be filtered."

[0033] In this disclosure, unless explicitly stated as a chroma block, the term "current block" can refer to a block including both a luma component block and a chroma component block, or to the "luma block of the current block." The luma component block of the current block may be explicitly expressed as a "luma block" or a "current luma block," including the explicit statement that it is a luma component block. Also, the chroma component block of the current block may be explicitly expressed as a "chroma block" or a "current chroma block," including the explicit statement that it is a chroma component block.

[0034] In the present disclosure, " / " and "," may be interpreted as "and / or." For example, "A / B" and "A, B" may be interpreted as "A and / or B." Also, "A / B / C" and "A, B, C" may mean "at least one of A, B, and / or C."

[0035] In this disclosure, "or" may be interpreted as "and / or." For example, "A or B" can mean 1) "A" only, 2) "B" only, or 3) "A and B." Alternatively, in this disclosure, "or" can mean "additionally or alternatively."

[0036] Video Coding System Overview

[0037] FIG. 1 is a schematic diagram illustrating a video coding system to which embodiments of the present disclosure can be applied.

[0038] A video coding system according to an embodiment may include an encoding device 10 and a decoding device 20. The encoding device 10 may transmit encoded video and / or image information or data to the decoding device 20 in a file or streaming format via a digital storage medium or a network.

[0039] An encoding device 10 according to an embodiment may include a video source generation unit 11, an encoding unit 12, and a transmission unit 13. A decoding device 20 according to an embodiment may include a reception unit 21, a decoding unit 22, and a rendering unit 23. The encoding unit 12 may be referred to as a video / video encoding unit, and the decoding unit 22 may be referred to as a video / video decoding unit. The transmission unit 13 may be included in the encoding unit 12. The reception unit 21 may be included in the decoding unit 22. The rendering unit 23 may include a display unit, which may be a separate device or an external component.

[0040] The video source generation unit 11 may acquire video / images through a process of capturing, synthesizing, or generating video / images. The video source generation unit 11 may include a video / image capture device and / or a video / image generation device. The video / image capture device may include, for example, one or more cameras, a video / image archive containing previously captured video / images, etc. The video / image generation device may include, for example, a computer, a tablet, a smartphone, etc., and may (electronically) generate video / images. For example, a virtual video / image may be generated by a computer, etc., in which case the video / image capture process may be replaced by a process of generating related data.

[0041] The encoder 12 may encode input video / image data. The encoder 12 may perform a series of procedures such as prediction, transformation, and quantization for compression and encoding efficiency. The encoder 12 may output encoded data (encoded video / image information) in the form of a bitstream.

[0042] The transmitter 13 may acquire encoded video / image information or data output in the form of a bitstream and transmit it to the receiver 21 of the decoding device 20 or another external object in the form of a file or streaming via a digital storage medium or a network. The digital storage medium may include various storage media such as USB, SD, CD, DVD, Blu-ray, HDD, and SSD. The transmitter 13 may include elements for generating a media file in a predetermined file format and elements for transmission via a broadcasting / communication network. The transmitter 13 may be provided as a transmission device separate from the encoder 12. In this case, the transmission device may include at least one processor for acquiring encoded video / image information or data output in the form of a bitstream and a transmitter for transmitting the same in the form of a file or streaming. The receiver 21 may extract / receive the bitstream from the storage medium or network and transmit it to the decoder 22.

[0043] The decoding unit 22 can decode the video / image by performing a series of procedures such as inverse quantization, inverse transformation, and prediction corresponding to the operations of the encoding unit 12.

[0044] The rendering unit 23 can render the decoded video / image, and the rendered video / image can be displayed on a display unit.

[0045] Overview of video encoding equipment

[0046] FIG. 2 is a schematic diagram illustrating a video encoding device to which an embodiment of the present disclosure can be applied.

[0047] 2, the video encoding device 100 may include a video division unit 110, a subtraction unit 115, a transform unit 120, a quantization unit 130, an inverse quantization unit 140, an inverse transform unit 150, an addition unit 155, a filtering unit 160, a memory 170, an inter prediction unit 180, an intra prediction unit 185, and an entropy encoding unit 190. The inter prediction unit 180 and the intra prediction unit 185 may be collectively referred to as a "prediction unit." The transform unit 120, the quantization unit 130, the inverse quantization unit 140, and the inverse transform unit 150 may be included in a residual processing unit. The residual processing unit may further include a subtraction unit 115.

[0048] Depending on the embodiment, all or at least some of the components constituting the video encoding device 100 may be implemented as a single hardware component (e.g., an encoder or a processor). Also, the memory 170 may include a decoded picture buffer (DPB) and may be implemented as a digital storage medium.

[0049] The video division unit 110 may divide an input video (or picture or frame) input to the video encoding device 100 into one or more processing units. For example, the processing units may be called coding units (CUs). The coding units may be obtained by recursively dividing a coding tree unit (CTU) or a largest coding unit (LCU) using a QT / BT / TT (quad-tree / binary-tree / ternary-tree) structure. For example, one coding unit may be divided into multiple coding units at deeper depths based on a quad-tree structure, a binary tree structure, and / or a ternary tree structure. To divide the coding units, a quad-tree structure may be applied first, and then a binary tree structure and / or a ternary tree structure may be applied later. The coding procedure according to the present disclosure may be performed based on the final coding unit that is not further divided. The largest coding unit may be directly used as the final coding unit, or a lower-depth coding unit obtained by dividing the largest coding unit may be used as the final coding unit. Here, the coding procedure may include procedures such as prediction, transformation, and / or reconstruction, which will be described later. As another example, a processing unit of the coding procedure may be a prediction unit (PU) or a transform unit (TU). The prediction unit and the transform unit may each be divided or partitioned from the final coding unit. The prediction unit may be a unit of sample prediction, and the transform unit may be a unit for deriving transform coefficients and / or a unit for deriving a residual signal from the transform coefficients.

[0050] The prediction unit (inter prediction unit 180 or intra prediction unit 185) may perform prediction on a current block (current block) to be processed and generate a predicted block including prediction samples for the current block. The prediction unit may determine whether intra prediction or inter prediction is applied to the current block or CU. The prediction unit may generate various information related to the prediction of the current block and transmit it to the entropy encoding unit 190. The prediction information may be encoded by the entropy encoding unit 190 and output in the form of a bitstream.

[0051] The intra prediction unit 185 may predict the current block by referring to samples in the current picture. The referenced samples may be located in the neighborhood of the current block or may be located far away from the current block depending on the intra prediction mode and / or intra prediction method. The intra prediction modes may include a plurality of non-directional modes and a plurality of directional modes. The non-directional modes may include, for example, a DC mode and a planar mode. The directional modes may include, for example, 33 directional prediction modes or 65 directional prediction modes depending on the accuracy of the prediction direction. However, this is merely an example, and more or less directional prediction modes may be used depending on the settings. The intra prediction unit 185 may also determine the prediction mode to be applied to the current block using the prediction modes applied to neighboring blocks.

[0052] The inter prediction unit 180 may derive a predicted block for a current block based on a reference block (reference sample array) identified by a motion vector in a reference picture. To reduce the amount of motion information transmitted in inter prediction mode, the motion information may be predicted in units of blocks, sub-blocks, or samples based on correlations between motion information of neighboring blocks and the current block. The motion information may include a motion vector and a reference picture index. The motion information may further include information on an inter prediction direction (e.g., L0 prediction, L1 prediction, or Bi prediction). In the case of inter prediction, the neighboring blocks may include spatial neighboring blocks present in the current picture and temporal neighboring blocks present in the reference picture. The reference picture including the reference block and the reference picture including the temporal neighboring blocks may be the same or different. The temporal neighboring blocks may be referred to as collocated reference blocks, collocated control units (colCUs), etc. The reference picture including the temporal neighboring blocks may be referred to as a collocated picture (colPic). For example, the inter predictor 180 may construct a motion information candidate list based on neighboring blocks and generate information indicating which candidate is used to derive a motion vector and / or a reference picture index for the current block. Inter prediction may be performed based on various prediction modes. For example, in the case of skip mode and merge mode, the inter predictor 180 may use motion information of neighboring blocks as motion information for the current block. In the case of skip mode, unlike in merge mode, a residual signal may not be transmitted.In the case of motion vector prediction (MVP) mode, the motion vector of a neighboring block is used as a motion vector predictor, and the motion vector of the current block can be signaled by encoding a motion vector difference and an indicator for the motion vector predictor. The motion vector difference can mean the difference between the motion vector of the current block and the motion vector predictor.

[0053] The predictor may generate a prediction signal based on various prediction methods and / or prediction techniques, which will be described later. For example, the predictor may apply intra prediction or inter prediction to predict the current block, or may simultaneously apply intra prediction and inter prediction. A prediction method that simultaneously applies intra prediction and inter prediction to predict the current block may be referred to as combined inter and intra prediction (CIIP). The predictor may also perform intra block copy (IBC) to predict the current block. Intra block copy may be used, for example, for coding content images / videos such as games, such as screen content coding (SCC). IBC is a method of predicting a current block using an already reconstructed reference block in a current picture that is located a predetermined distance away from the current block. When IBC is applied, the position of the reference block in the current picture may be coded as a vector (block vector) corresponding to the predetermined distance. IBC basically performs prediction within the current picture, but may be performed similarly to inter prediction in that a reference block is derived within the current picture. That is, IBC can use at least one of the inter prediction techniques described in this disclosure.

[0054] The prediction signal generated by the prediction unit may be used to generate a restored signal or a residual signal. The subtraction unit 115 may subtract the prediction signal (predicted block, predicted sample array) output from the prediction unit from the input video signal (original block, original sample array) to generate a residual signal (residual signal, residual block, residual sample array). The generated residual signal may be transmitted to the conversion unit 120.

[0055] The transform unit 120 may generate transform coefficients by applying a transform technique to the residual signal. For example, the transform technique may include at least one of a Discrete Cosine Transform (DCT), a Discrete Sine Transform (DST), a Karhunen-Loeve Transform (KLT), a Graph-Based Transform (GBT), or a Conditionally Non-linear Transform (CNT). Here, the GBT refers to a transform obtained from a graph when relationship information between pixels is expressed as a graph. The CNT refers to a transform obtained based on a predicted signal generated using all previously reconstructed pixels. The transform process may be applied to pixel blocks having the same square size or to blocks of variable size other than a square.

[0056] The quantization unit 130 may quantize the transform coefficients and transmit the quantized transform coefficients to the entropy encoding unit 190. The entropy encoding unit 190 may encode the quantized signal (information about the quantized transform coefficients) and output it as a bitstream. The information about the quantized transform coefficients may be referred to as residual information. The quantization unit 130 may rearrange the quantized transform coefficients in a block form into a one-dimensional vector form based on a coefficient scan order, and may generate information about the quantized transform coefficients based on the quantized transform coefficients in the one-dimensional vector form.

[0057] The entropy encoding unit 190 may perform various encoding methods, such as exponential Golomb, context-adaptive variable length coding (CAVLC), context-adaptive binary arithmetic coding (CABAC), etc. The entropy encoding unit 190 may encode information required for video / image restoration (e.g., values ​​of syntax elements) together with or separately from the quantized transform coefficients. The encoded information (e.g., encoded video / video information) may be transmitted or stored in the form of a bitstream in network abstraction layer (NAL) units. The video / video information may further include information on various parameter sets, such as an adaptation parameter set (APS), a picture parameter set (PPS), a sequence parameter set (SPS), or a video parameter set (VPS). The video / video information may also include general constraint information. The signaling information, transmitted information, and / or syntax elements referred to in this disclosure may be encoded according to the encoding procedures described above and included in the bitstream.

[0058] The bitstream may be transmitted over a network or stored in a digital storage medium. Here, the network may include a broadcasting network and / or a communication network, and the digital storage medium may include various storage media such as a USB, SD, CD, DVD, Blu-ray, HDD, SSD, etc. A transmitter (not shown) for transmitting and / or a storage unit (not shown) for storing the signal output from the entropy encoding unit 190 may be provided as an internal / external element of the video encoding device 100, or the transmitter may be provided as a component of the entropy encoding unit 190.

[0059] The quantized transform coefficients output from the quantization unit 130 may be used to generate a residual signal. For example, the quantized transform coefficients may be subjected to inverse quantization and inverse transformation in the inverse quantization unit 140 and the inverse transform unit 150, respectively, to reconstruct a residual signal (residual block or residual sample).

[0060] The adder 155 may generate a reconstructed signal (reconstructed picture, reconstructed block, reconstructed sample array) by adding the reconstructed residual signal to a prediction signal output from the inter prediction unit 180 or the intra prediction unit 185. When there is no residual for the current block to be processed, such as when a skip mode is applied, the predicted block may be used as the reconstructed block. The adder 155 may be referred to as a reconstruction unit or a reconstructed block generation unit. The generated reconstructed signal may be used for intra prediction of the next current block to be processed in the current picture, or may be used for inter prediction of the next picture after filtering, as described below.

[0061] The filtering unit 160 may apply filtering to the reconstructed signal to improve subjective / objective image quality. For example, the filtering unit 160 may apply various filtering methods to the reconstructed picture to generate a modified reconstructed picture and store the modified reconstructed picture in the memory 170, specifically, in the DPB of the memory 170. The various filtering methods may include, for example, deblocking filtering, sample adaptive offset, an adaptive loop filter, a bilateral filter, etc. The filtering unit 160 may generate various information related to filtering and transmit it to the entropy encoding unit 190, as will be described later in the description of each filtering method. The filtering information may be encoded by the entropy encoding unit 190 and output in the form of a bitstream.

[0062] The modified reconstructed picture transmitted to the memory 170 may be used as a reference picture in the inter prediction unit 180. This allows the video encoding device 100 to avoid prediction mismatch between the video encoding device 100 and the video decoding device when inter prediction is applied, and also improves encoding efficiency.

[0063] The DPB in the memory 170 may store a modified reconstructed picture to be used as a reference picture in the inter prediction unit 180. The memory 170 may store motion information of a block from which motion information in the current picture is derived (or encoded) and / or motion information of a block in an already reconstructed picture. The stored motion information may be transmitted to the inter prediction unit 180 to be used as motion information of a spatially neighboring block or a temporally neighboring block. The memory 170 may store reconstructed samples of reconstructed blocks in the current picture and transmit them to the intra prediction unit 185.

[0064] Overview of the video decoder

[0065] FIG. 3 is a schematic diagram illustrating a video decoding device to which an embodiment of the present disclosure can be applied.

[0066] 3, the video decoding apparatus 200 may include an entropy decoding unit 210, an inverse quantization unit 220, an inverse transform unit 230, an adder 235, a filtering unit 240, a memory 250, an inter prediction unit 260, and an intra prediction unit 265. The inter prediction unit 260 and the intra prediction unit 265 may be collectively referred to as a "prediction unit." The inverse quantization unit 220 and the inverse transform unit 230 may be included in a residual processing unit.

[0067] Depending on the embodiment, all or at least some of the components constituting the video decoding device 200 may be implemented as a single hardware component (e.g., a decoder or a processor). Also, the memory 170 may include a DPB and may be implemented as a digital storage medium.

[0068] The video decoding apparatus 200, which receives a bitstream including video / image information, may reconstruct an image by performing a process corresponding to the process performed by the video encoding apparatus 100 of FIG. 2. For example, the video decoding apparatus 200 may perform decoding using a processing unit applied in the video encoding apparatus. Accordingly, the decoding processing unit may be, for example, a coding unit. The coding unit may be a coding tree unit or may be obtained by dividing a maximum coding unit. The reconstructed video signal decoded and output by the video decoding apparatus 200 may be reproduced by a playback device (not shown).

[0069] The video decoding apparatus 200 may receive a signal output from the video encoding apparatus of FIG. 2 in the form of a bitstream. The received signal may be decoded by the entropy decoding unit 210. For example, the entropy decoding unit 210 may parse the bitstream to derive information (e.g., video / video information) necessary for video restoration (or picture restoration). The video / video information may further include information on various parameter sets, such as an adaptation parameter set (APS), a picture parameter set (PPS), a sequence parameter set (SPS), or a video parameter set (VPS). The video / video information may also include general constraint information. The video decoding apparatus may further use the information on the parameter sets and / or the general constraint information to decode the video. Signaling information, received information, and / or syntax elements referred to in this disclosure may be obtained from the bitstream by being decoded by the decoding procedure. For example, the entropy decoding unit 210 may decode information in a bitstream based on a coding method such as Exponential-Golomb coding, CAVLC, or CABAC, and output values ​​of syntax elements required for image restoration and quantized values ​​of transform coefficients related to residuals. More specifically, the CABAC entropy decoding method receives bins corresponding to each syntax element in the bitstream, determines a context model using information on the syntax element to be decoded, decoding information on neighboring blocks and the block to be decoded, or information on symbols / bins decoded in a previous step, predicts the occurrence probability of bins according to the determined context model, and performs arithmetic decoding of the bins to generate symbols corresponding to the values ​​of each syntax element.In this case, after determining a context model, the CABAC entropy decoding method may update the context model using information about the decoded symbol / bin for the context model of the next symbol / bin. Prediction information from the information decoded by the entropy decoding unit 210 may be provided to a prediction unit (the inter prediction unit 260 and the intra prediction unit 265), and residual values ​​entropy decoded by the entropy decoding unit 210, i.e., quantized transform coefficients and related parameter information, may be input to the inverse quantization unit 220. In addition, filtering information from the information decoded by the entropy decoding unit 210 may be provided to the filtering unit 240. Meanwhile, a receiving unit (not shown) for receiving a signal output from the video encoding device may be further provided as an internal / external element of the video decoding device 200, or the receiving unit may be provided as a component of the entropy decoding unit 210.

[0070] Meanwhile, a video decoding apparatus according to the present disclosure may be referred to as a video / image / picture decoding apparatus. The video decoding apparatus may include an information decoder (video / image / picture information decoder) and / or a sample decoder (video / image / picture sample decoder). The information decoder may include an entropy decoding unit 210, and the sample decoder may include at least one of an inverse quantization unit 220, an inverse transform unit 230, an adder 235, a filtering unit 240, a memory 250, an inter prediction unit 260, and an intra prediction unit 265.

[0071] The inverse quantization unit 220 may inverse quantize the quantized transform coefficients and output the transform coefficients. The inverse quantization unit 220 may rearrange the quantized transform coefficients in a two-dimensional block format. In this case, the rearrangement may be performed based on the coefficient scanning order performed in the video encoding device. The inverse quantization unit 220 may inverse quantize the quantized transform coefficients using a quantization parameter (e.g., quantization step size information) to obtain transform coefficients.

[0072] The inverse transform unit 230 can inversely transform the transform coefficients to obtain a residual signal (residual block, residual sample array).

[0073] The prediction unit may perform prediction on a current block and generate a predicted block including prediction samples for the current block. The prediction unit may determine whether intra prediction or inter prediction is applied to the current block based on the prediction information output from the entropy decoding unit 210, and may determine a specific intra / inter prediction mode (prediction method).

[0074] As mentioned in the description of the prediction unit of the video encoding device 100, the prediction unit can generate a prediction signal based on various prediction methods (techniques) described below.

[0075] The intra predictor 265 may predict the current block by referring to samples in the current picture. The description of the intra predictor 185 may also be applied to the intra predictor 265.

[0076] The inter prediction unit 260 may derive a predicted block for the current block based on a reference block (reference sample array) identified by a motion vector in a reference picture. To reduce the amount of motion information transmitted in inter prediction mode, the motion information may be predicted in units of blocks, sub-blocks, or samples based on correlations between motion information of neighboring blocks and the current block. The motion information may include a motion vector and a reference picture index. The motion information may further include information on an inter prediction direction (e.g., L0 prediction, L1 prediction, Bi prediction, etc.). In the case of inter prediction, the neighboring blocks may include spatial neighboring blocks in the current picture and temporal neighboring blocks in the reference picture. For example, the inter prediction unit 260 may construct a motion information candidate list based on the neighboring blocks and derive a motion vector and / or a reference picture index for the current block based on received candidate selection information. Inter prediction may be performed based on various prediction modes (methods), and the prediction information may include information indicating the inter prediction mode (method) for the current block.

[0077] The adder 235 may generate a reconstructed signal (reconstructed picture, reconstructed block, reconstructed sample array) by adding the obtained residual signal to a prediction signal (predicted block, predicted sample array) output from a prediction unit (including the inter prediction unit 260 and / or the intra prediction unit 265). When there is no residual for the current block to be processed, such as when a skip mode is applied, the predicted block may be used as the reconstructed block. The description of the adder 155 may also apply to the adder 235. The adder 235 may be referred to as a reconstruction unit or a reconstructed block generation unit. The generated reconstructed signal may be used for intra prediction of the next current block to be processed in the current picture, or may be used for inter prediction of the next picture after undergoing filtering, as described below.

[0078] The filtering unit 240 may apply filtering to the reconstructed signal to improve subjective / objective image quality. For example, the filtering unit 240 may apply various filtering methods to the reconstructed picture to generate a modified reconstructed picture, and may store the modified reconstructed picture in the memory 250, specifically, in a DPB of the memory 250. The various filtering methods may include, for example, deblocking filtering, sample adaptive offset, an adaptive loop filter, a bilateral filter, etc.

[0079] The (modified) reconstructed picture stored in the DPB of the memory 250 may be used as a reference picture in the inter prediction unit 260. The memory 250 may store motion information of a block from which motion information in the current picture is derived (or decoded) and / or motion information of a block in an already reconstructed picture. The stored motion information may be transmitted to the inter prediction unit 260 to be used as motion information of a spatially neighboring block or a temporally neighboring block. The memory 250 may store reconstructed samples of reconstructed blocks in the current picture and transmit them to the intra prediction unit 265.

[0080] In this specification, the embodiments described for the filtering unit 160, inter prediction unit 180, and intra prediction unit 185 of the video encoding device 100 may also be applied identically or correspondingly to the filtering unit 240, inter prediction unit 260, and intra prediction unit 265 of the video decoding device 200, respectively.

[0081] Neural-network post-filter characteristics (NNPFC)

[0082] The combination of Tables 1 and 2 represents the NNPFC syntax structure.

[0083] [Table 1]

[0084] [Table 2]

[0085] The NNPFC syntax structures of Tables 1 and 2 may be signaled in the form of a supplemental enhancement information (SEI) message. An SEI message that signals the NNPFC syntax structures of Tables 1 and 2 may be referred to as an NNPFC SEI message.

[0086] The NNPFC SEI message can specify a neural network that can be used as a post-processing filter. The use of a specified post-processing filter for a particular picture can be indicated using neural-network post-filter activation SEI messages. Here, "post-processing filter" and "post-filter" may have the same meaning.

[0087] To use such an SEI message, it may be necessary to define variables such as:

[0088] The width and height of the decoded output picture may be cropped in luma samples, and this width and height can be indicated by CroppedWidth and CroppedHeight, respectively.

[0089] - The luma sample array CroppedYPic[idx] and chroma sample arrays CroppedCbPic[idx] and CroppedCrPic[idx] of the cropped decoded output picture may be used as inputs to the post-processing filter if present, where idx may range from 0 to numInputPics-1.

[0090] - BitDepth Y may indicate the bit depth for the luma sample array of the cropped decoded output picture.

[0091] - BitDepth C may indicate the bit depth of the chroma sample array (if any) of the cropped decoded output picture.

[0092] - ChromaFormatIdc may indicate a chroma format identifier.

[0093] - When the value of nnpfc_auxiliary_inp_idc is 1, the filtering strength control value StrengthControlVal must be a real number in the range of 0 to 1.

[0094] The variables SubWidthC and SubHeightC may be derived from ChromaFormatIdc. Two or more NNPFC SEI messages may exist for the same picture. When two or more NNPFC SEI messages with different nnpfc_id values ​​exist or are activated for the same picture, the two or more NNPFC SEI messages may have the same or different nnpfc_purpose and nnpfc_mode_idx values.

[0095] nnpfc_id may contain an identification number that can be used to identify a post-processing filter. The nnpfc_id value is 0 to 2. 32Must be in the range 256 to 511 and 2 31 ~2 32 nnpfc_id values ​​in the -2 range may be reserved for future use. Decoders may use nnpfc_id values ​​in the 256-511 range or 2 31 ~2 32 NNPFC SEI messages with nnpfc_id in the -2 range shall be ignored.

[0096] If the NNPFC SEI message is the first NNPFC SEI message in decoding order with a particular nnpfc_id value currently in the CLVS, the following may apply.

[0097] The SEI message may indicate a base post-processing filter.

[0098] The SEI message may relate to the currently decoded picture and all subsequent decoded pictures of the current layer in output order until the current CLVS is finished.

[0099] An NNPFC SEI message may be a repetition of a previous NNPFC SEI message currently in the CLVS in decoding order, and the subsequent semantics may apply as if this SEI message were the only NNPFC SEI message with the same content currently in the CLVS.

[0100] If the NNPFC SEI message is not the first NNPFC SEI message in decoding order with a particular nnpfc_id value currently in the CLVS, the following may apply.

[0101] - The SEI message may be associated in output order with the current CLVS or the currently decoded picture and all subsequent decoded pictures of the current layer until the current CLVS is terminated, or may be associated with the next NNPFC SEI message having a specific nnpfc_id value in the current CLVS in output order.

[0102] If the NNPFC SEI message is the first NNPFC SEI message in decoding order currently having a particular nnpfc_id value within the CLVS, a value of 1 in nnpfc_mode_idc may indicate that the basic post-processing filter associated with the nnpfc_id value is a neural network, which may be a neural network identified by a URI indicated in nnpfc_uri using the format identified in the tag URI nnpfc_tag_uri.

[0103] If the NNPFC SEI message is not the first NNPFC SEI message in decoding order with a particular nnpfc_id value currently in the CLVS, a value of 1 in nnpfc_mode_idc may indicate that updates to basic post-processing filters with the same nnpfc_id value are defined by the URI indicated in nnpfc_uri using the format identified in the tag URI nnpfc_tag_uri.

[0104] The value of nnpfc_mode_idc may be restricted to have a range of 0 to 1 in the bitstream. Values ​​in the range 2 to 255 for nnpfc_mode_idc may be reserved for future use and may not be present in the bitstream. Decoders MUST ignore NNPFC SEI messages with nnpfc_mode_idc in the range 2 to 255. Values ​​of nnpfc_mode_idc greater than 255 may not be present in the bitstream and may not be reserved for future use.

[0105] If the SEI message is the first NNPFC SEI message in decoding order with a specific nnpfc_id value in the current CLVS, the post-processing filter PostProcessingFilter() may be assigned the same as the basic post-processing filter.

[0106] If the SEI message is not the first NNPFC SEI message in decoding order with a particular nnpfc_id value currently in the CLVS, the post-processing filter PostProcessingFilter() may be obtained by applying the updates defined by the SEI message to the basic post-processing filter.

[0107] Updates are not cumulative; rather, each update may be applied to the basic post-processing filter, which is the post-processing filter specified by the first NNPFC SEI message in decoding order with a particular nnpfc_id value currently in the CLVS.

[0108] nnpfc_reserved_zero_bit_a may be constrained by bitstream restrictions to have a value equal to 0. Decoders may be constrained to ignore NNPFC SEI messages where the value of nnpfc_reserved_zero_bit_a is not 0.

[0109] The nnpfc_tag_uri may contain a tag URI with syntax and semantics specified in IETF RFC 4151 that identifies a neural network to be used as a base post-processing filter or an update to the base post-processing filter using the nnpfc_id value identified by the nnpfc_uri. The nnpfc_tag_uri can be used to uniquely identify the type of neural network data specified by the nnrpf_uri without a central registration authority. An nnpfc_tag_uri equal to "tag:iso.org,2023:15938-17" can indicate that the neural network data identified by the nnpfc_uri complies with ISO / IEC 15938-17.

[0110] The nnpfc_uri may contain a URI with syntax and semantics specified in IETF Internet Standard 66 that identifies the neural network used as the base post-processing filter or an update to the base post-processing filter that uses the same nnpfc_id value.

[0111] A value of 1 for nnpfc_formatting_and_purpose_flag may indicate the presence of syntax elements related to the filter's purpose, input formatting, output formatting, and complexity. A value of 0 for nnpfc_formatting_and_purpose_flag may indicate the absence of syntax elements related to the filter's purpose, input formatting, output formatting, and complexity.

[0112] If the SEI message is the first NNPFC SEI message with a specific nnpfc_id value in the current CLVS in decoding order, the value of nnpfc_formatting_and_purpose_flag shall be equal to 1. If the SEI message is not the first NNPFC SEI message with a specific nnpfc_id value in the current CLVS in decoding order, the value of nnpfc_formatting_and_purpose_flag shall be equal to 0.

[0113] nnpfc_purpose can indicate the purpose of the post-processing filter specified in Table 3.

[0114] Values ​​of nnpfc_purpose MUST be in the range 0 to 5 due to bitstream restrictions. Values ​​6 to 1023 for nnpfc_purpose MAY not be present in the bitstream and are reserved for future use. Decoders MUST ignore NNPFC SEI messages with nnpfc_purpose in the range 6 to 1203. nnpfc_purpose values ​​greater than 1023 MAY not be present in the bitstream and are not reserved for future use.

[0115] [Table 3]

[0116] When a reserved value of nnpfc_purpose is later used, the syntax of this SEI message may be extended to existing syntax elements, provided that nnpfc_purpose is identical to that value.

[0117] If SubWidthC has a value of 1 and SubHeightC has a value of 1, then nnpfc_purpose must not have a value of 2 or 4.

[0118] A value of 1 for nnpfc_out_sub_c_flag may indicate that outSubWidthC has a value of 1 and that outSubHeightC has a value of 1. A value of 0 for nnpfc_out_sub_c_flag may indicate that outSubWidthC has a value of 2 and that outSubHeightC has a value of 1. If nnpfc_out_sub_c_flag is not present, outSubWidthC may be inferred to be the same as SubWidthC and outSubHeightC may be inferred to be the same as SubHeightC. If ChromaFormatIdc has a value of 2 and nnpfc_out_sub_c_flag is present, the value of nnpfc_out_sub_c_flag must be equal to 1.

[0119] nnpfc_pic_width_in_luma_samples and nnpfc_pic_height_in_luma_samples may indicate the width and height, respectively, of the picture's luma sample array that results from applying the post-processing filter identified by nnpfc_id to the cropped decoded output picture. If nnpfc_pic_width_in_luma_samples and nnpfc_pic_height_in_luma_samples are not present, they may be inferred to be the same as CroppedWidth and CroppedHeight, respectively. The value of nnpfc_pic_width_in_luma_samples should be in the range of CroppedWidth to CroppedWidth*16-1. The value of nnpfc_pic_height_in_luma_samples should be in the range of CroppedHeight to CroppedHeight*16-1.

[0120] nnpfc_num_input_pics_minus2+2 may indicate the number of decoded output pictures to be used as input for the post-processing filter.

[0121] nnpfc_interpolated_pics[i] may indicate the number of interpolated pictures generated by the post-processing filter between the i-th picture and the (i+1)-th picture that are used as input to the post-processing filter.

[0122] The variable numInputPics indicating the number of pictures used as input for the post-processing filter and the variable numOutputPics indicating the total number of pictures generated as a result of the post-processing filter may be derived as shown in Table 4.

[0123] [Table 4]

[0124] A value of 1 for nnpfc_component_last_flag can indicate that the last dimension of the input tensor inputTensor for the post-processing filter and the output tensor outputTensor that is the result of the post-processing filter are currently used for the channel. A value of 0 for nnpfc_component_last_flag can indicate that the third dimension of the input tensor inputTensor for the post-processing filter and the output tensor outputTensor that is the result of the post-processing filter are currently used for the channel.

[0125] The first dimension of the input tensor and output tensor may be used as a batch index used in some neural network frameworks. Although the formula in the semantics of this SEI message uses a batch size corresponding to a batch index such as 0, determining the batch size used as input for neural network inference may be determined by the implementation of post-processing.

[0126] For example, when the value of nnpfc_inp_order_idc is equal to 3 and the value of nnpfc_auxiliary_inp_idc is equal to 1, the input tensor may have seven channels, including four luma matrices, two chroma matrices, and one auxiliary input matrix. In this case, the DeriveInputTensors() process may derive each of the seven channels of the input tensor one by one, and when a particular channel among these channels is processed, that channel may be referred to as the current channel during the process.

[0127] nnpfc_inp_format_idc may indicate how to convert sample values ​​of the cropped decoded output picture to input values ​​of the post-processing filter. If nnpfc_inp_format_idc is 0, the input values ​​for the post-processing filter are real numbers, and the InpY() and InpC() functions may be specified as in Equation 1.

[0128]

number

[0129] If the value of nnpfc_inp_format_idc is 1, the input values ​​of the post-processing filter are unsigned integer numbers, and the InpY() and InpC() functions may be derived as shown in Table 5.

[0130] [Table 5]

[0131] The variable inpTensorBitDepth may be derived from the syntax element nnpfc_inp_tensor_bitlength_minus8 described below.

[0132] Values ​​of nnpfc_inp_format_idc greater than 1 may be reserved for future use and may not be present in the bitstream. Decoders MUST ignore NNPFC SEI messages that contain reserved values ​​of nnpfc_inp_format_idc.

[0133] nnpfc_inp_tensor_bitlength_minus8+8 may indicate the bit depth of the luma sample values ​​in the input integer tensor. The value of inpTensorBitDepth may be derived as in Equation 2.

[0134]

number

[0135] The value of nnpfc_inp_tensor_bitlength_minus8 may be restricted to lie in the range 0 to 24.

[0136] nnpfc_inp_order_idc can indicate how to align the sample array of the cropped decoded output picture to one of the input pictures for the post-processing filter.

[0137] Values ​​of nnpfc_inp_order_idc MUST lie in the range 0 to 3 in the bitstream. Values ​​of 4 to 255 for nnpfc_inp_order_idc MUST NOT lie in the bitstream. Decoders MUST ignore NNPFC SEI messages with nnpfc_inp_order_idc lying in the range 4 to 255. Values ​​of nnpfc_inp_order_idc greater than 255 MUST NOT lie in the bitstream and are not reserved for future use.

[0138] If the value of ChromaFormatIdc is not 1, the value of nnpfc_inp_order_idc must not be 3.

[0139] Table 6 contains descriptions for the nnpfc_inp_order_idc values.

[0140] [Table 6] JPEG2026502277000010.jpg44167

[0141] A patch may be a rectangular array of samples from a component of a picture (eg, luma or chroma component).

[0142] An nnpfc_auxiliary_inp_idc greater than 0 may indicate that auxiliary input data is present in the input tensor of the neural network post-filter. A value of 0 for nnpfc_auxiliary_inp_idc may indicate that auxiliary input data is not present in the input tensor. A value of 1 for nnpfc_auxiliary_inp_idc may indicate that auxiliary input data is derived by the methods disclosed in Tables 7-9.

[0143] Values ​​of nnpfc_auxiliary_inp_idc MUST be in the range 0 to 1 in the bitstream. Values ​​2 to 255 for nnpfc_inp_order_idc MUST NOT be present in the bitstream. Decoders MUST ignore NNPFC SEI messages with nnpfc_inp_order_idc in the range 2 to 255. Values ​​of nnpfc_inp_order_idc greater than 255 MUST NOT be present in the bitstream and are not reserved for future use.

[0144] The process DeriveInputTensors() for deriving the input tensor inputTensor for a given vertical sample coordinate cTop and a horizontal sample coordinate cLeft that specifies the top left sample position of the sample patch included in the input tensor can be shown as a combination of Tables 7 to 9.

[0145] [Table 7]

[0146] [Table 8]

[0147] [Table 9]

[0148] A value of 1 for nnpfc_separate_colour_description_present_flag can indicate that the unique combination of colour primaries, transformation characteristics, and matrix coefficients for the post-processing filter picture is specified in the SEI message syntax structure. A value of 0 for nnfpc_separate_colour_description_present_flag can indicate that the combination of colour primaries, transformation characteristics, and matrix coefficients for the post-processing filter picture is the same as that displayed in the CLVS VUI parameters.

[0149] nnpfc_colour_primaries may have the same semantics as defined for the vui_colour_primaries syntax element, except as follows:

[0150] - nnpfc_colour_primaries can indicate the primary colors of the picture that appear as a result of applying the neural network postfilter specified in the SEI message, rather than the primary colors used in CLVS.

[0151] - If nnpfc_colour_primaries is not present in the NNPFC SEI message, the value of nnpfc_colour_primaries may be inferred to be the same as the value of vui_colour_primaries.

[0152] nnpfc_transfer_characteristics may have the same semantics as defined for the vui_transfer_characteristics syntax element, except as follows:

[0153] - nnpfc_transfer_characteristics can indicate the transfer characteristics of the picture that appears as a result of applying the neural network post-filter specified in the SEI message, rather than the transfer characteristics used in CLVS.

[0154] - If nnpfc_transfer_characteristics is not present in the NNPFC SEI message, the value of nnpfc_transfer_characteristics may be inferred to be the same as the value of vui_transfer_characteristics.

[0155] nnpfc_matrix_coeffs may have the same semantics as specified for the vui_matrix_coeffs syntax element, except as follows:

[0156] - nnpfc_matrix_coeffs can indicate the matrix coefficients of the picture that appears as a result of applying the neural network post-filter specified in the SEI message, rather than the matrix coefficients used for CLVS.

[0157] - If nnpfc_matrix_coeffs is not present in the NNPFC SEI message, the value of nnpfc_matrix_coeffs may be inferred to be the same as the value of vui_matrix_coeffs.

[0158] - The values ​​allowed for nnpfc_matrix_coeffs may not be restricted by the chroma format of the decoded video picture indicated by the ChromaFormatIdc value for the semantics of the VUI parameter.

[0159] - If the value of nnpfc_matrix_coeffs is equal to 0, the value of nnpfc_out_order_idc must not be equal to 1 or 3.

[0160] The value 0 of nnpfc_out_format_id indicates that for the bit depth bitDepth required for subsequent post-processing or display, the sample values output by the post-processing filter are real numbers that are linearly mapped from the range of values from 0 to 1 to the range of unsigned integer values from 0 to (1<<bitDepth)-1. The value 1 of nnpfc_out_format_flag can indicate that the sample values output by the post-processing filter are unsigned integers in the range of 0 to (1<<(nnpfc_out_tensor_bitlength_minus8+8))-1. Values of nnpfc_out_format_idc greater than 1 do not exist in the bitstream. The decoder must ignore NNPFC SEI messages containing reserved values of nnpfc_out_format_idc. "+8" can indicate the bit depth of the sample values in the output integer tensor. The value of nnpfc_out_tensor_bitlength_minus8 must exist in the range of 0 to 24.

[0161] nnpfc_out_order_idc can indicate the output order of the samples output from the post-processing filter. The value of nnpfc_out_order_idc must exist in the range of 0 to 3 in the bitstream. Values from 4 to 255 for nnpfc_out_order_idc do not exist in the bitstream. The decoder must ignore NNPFC SEI messages having nnpfc_out_order_idc in the range of 4 to 255. Values of nnpfc_out_order_idc greater than 255 do not exist in the bitstream and are not reserved for future use. When the value of nnpfc_purpose is 2 or 4, the value of nnpfc_out_order_idc must not be the same as 3.

[0162] Table 10 shows the explanations for the values of nnpfc_out_order_idc.

[0163] [Table 10]

[0164] The process StoreOutputTensors() for deriving sample values ​​in filtered output sample arrays FilteredYPic, FilteredCbPic, and FilteredCrPic from the output tensor outputTensor for a given vertical sample coordinate cTop and horizontal sample coordinate cLeft indicating the top left sample position for the patch of samples contained in the input tensor may be expressed as a combination of Table 11 and Table 12.

[0165] [Table 11]

[0166] [Table 12]

[0167] A value of 1 for nnpfc_constant_patch_size_flag can indicate that the post-processing filter accepts as input exactly the patch size indicated by nnpfc_patch_width_minus1 and nnpfc_patch_height_minus1. A value of 0 for nnpfc_constant_patch_size_flag can indicate that the post-processing filter accepts as input all patch sizes that are positive integer multiples of the patch size indicated by nnpfc_patch_width_minus1 and nnpfc_patch_height_minus1.

[0168] npfc_patch_width_minus1+1 can indicate the number of horizontal samples of the patch size required for input to the post-processing filter when the value of nnpfc_constant_patch_size_flag is 1. The value of nnpfc_patch_width_minus1 must be in the range of 0 to Min(32766,CroppedWidth-1).

[0169] npfc_patch_height_minus1+1 can indicate the number of vertical samples of the patch size required for input to the post-processing filter when the value of nnpfc_constant_patch_size_flag is 1. The value of nnpfc_patch_height_minus1 must be in the range of 0 to Min(32766,CroppedHeight-1).

[0170] The inpPatchWidth and inpPatchHeight variables may be set to the patch size width and patch size height, respectively.

[0171] If the value of nnpfc_constant_patch_size_flag is 0, the following may apply:

[0172] The values ​​of inpPatchWidth and inpPatchHeight may be provided by external means or set by the post-processor itself.

[0173] - The value of inpPatchWidth must be a positive integer multiple of nnpfc_patch_width_minus1+1 and must be less than or equal to CroppedWidth. The value of inpPatchHeight must be a positive integer multiple of nnpfc_patch_height_minus1+1 and must be less than or equal to CroppedHeight.

[0174] Otherwise (if the value of nnpfc_constant_patch_size_flag is 1), the value of inpPatchWidth may be set equal to nnpfc_patch_width_minus1+1, and the value of inpPatchHeight may be set equal to nnpfc_patch_height_minus1+1.

[0175] nnpfc_overlap can indicate the number of overlapping horizontal and vertical samples of adjacent input tensors of the post-processing filter. The value of nnpfc_overlap must be in the range of 0 to 16383.

[0176] The variables outPatchWidth, outPatchHeight, horCScaling, verCScaling, outPatchCWidth, outPatchCHeight, and overlapSize may be derived as shown in Table 13.

[0177] [Table 13]

[0178] It is a bitstream conformance requirement that outPatchWidth*CroppedWidth must be equal to nnpfc_pic_width_in_luma_samples*inpPatchWidth and outPatchHeight*CroppedHeight must be equal to nnpfc_pic_height_in_luma_samples*inpPatchHeight.

[0179] nnpfc_padding_type can indicate the padding process when referencing sample positions that are outside the boundaries of the cropped decoded output picture, as described in Table 14. The value of nnpfc_padding_type must be in the range 0 to 15.

[0180] [Table 14]

[0181] nnpfc_luma_padding_val can indicate the luma value to use for padding when the value of nnpfc_padding_type is 4.

[0182] nnpfc_cb_padding_val can indicate the Cb value to be used for padding when the value of nnpfc_padding_type is 4.

[0183] nnpfc_cr_padding_val can indicate the Cr value to be used for padding when the value of nnpfc_padding_type is 4.

[0184] The InpSampleVal(y, x, picHeight, picWidth, CroppedPic) function, whose inputs are vertical sample position y, horizontal sample position x, picture height picHeight, picture width picWidth, and sample array CroppedPic, can return the derived value of SampleVal as shown in Table 15.

[0185] For inputs to the InpSampleVal() function, the vertical positions may be listed before the horizontal positions for compatibility with the input tensor conventions of some inference engines.

[0186] [Table 15]

[0187] The process in Table 16 may be used to patch-wise filter the cropped decoded output picture using the post-processing filter PostProcessingFilter() to generate a filtered picture, which may include a Y sample array FilteredYPic, a Cb sample array FilteredCbPic, and a Cr sample array FilteredCrPic, as indicated by nnpfc_out_order_idc.

[0188] [Table 16]

[0189] A value of 1 for nnpfc_complexity_info_present_flag may indicate the presence of one or more syntax elements indicating the complexity of the post-processing filter associated with nnpfc_id. A value of 0 for nnpfc_complexity_info_present_flag may indicate the absence of syntax elements indicating the complexity of the post-processing filter associated with nnpfc_id.

[0190] A value of 0 for nnpfc_parameter_type_idc may indicate that the neural network uses integer parameters only. A value of 1 for nnpfc_parameter_type_flag may indicate that the neural network can use floating-point or integer parameters. A value of 2 for nnpfc_parameter_type_idc may indicate that the neural network uses binary parameters only. A value of 3 for nnpfc_parameter_type_idc may be reserved for future use and will not be present in the bitstream. Decoders must ignore NNPFC SEI messages with a value of 3 for nnpfc_parameter_type_idc.

[0191] Values ​​0, 1, 2, and 3 of nnpfc_log2_parameter_bit_length_minus3 can indicate that the neural network shall not use parameters with bit lengths greater than 8, 16, 32, and 64, respectively. If nnpfc_parameter_type_idc is present and nnpfc_log2_parameter_bit_length_minus3 is not present, the neural network shall not use parameters with bit lengths greater than 1.

[0192] nnpfc_num_parameters_idc can indicate the maximum number of neural network parameters for the post-processing filter in powers of 2048. A value of 0 for nnpfc_num_parameters_idc can indicate that the maximum number of neural network parameters is unknown. Values ​​of nnpfc_num_parameters_idc must be in the range 0 to 52. Values ​​of nnpfc_num_parameters_idc greater than 52 shall not be present in the bitstream. Decoders must ignore NNPFC SEI messages with nnpfc_num_parameters_idc greater than 52.

[0193] If the value of nnpfc_num_parameters_idc is greater than 0, the maxNumParameters variable may be derived as shown in Equation 3.

[0194]

number

[0195] The number of neural network parameters in a post-processing filter may be limited to a number less than or equal to maxNumParameters.

[0196] A value of nnpfc_num_kmac_operations_idc greater than 0 can indicate that the maximum number of multiply-accumulate operations per sample of the post-processing filter is less than or equal to nnpfc_num_kmac_operations_idc * 1000. A value of 0 for nnpfc_num_kmac_operations_idc can indicate that the maximum number of multiply-accumulate operations for the network is unknown. The value of nnpfc_num_kmac_operations_idc can be between 0 and 2. 32 Must be in the range -1.

[0197] A value of nnpfc_total_kilobyte_size greater than 0 can indicate the total size (in kilobytes) required to store the uncompressed parameters of the neural network. The total size in bits can be a number greater than or equal to the sum of the bits used to store each parameter. nnpfc_total_kilobyte_size can be the total size (in bits) divided by 8000 and rounded up or down. A value of 0 for nnpfc_total_kilobyte_size can indicate that the total size required to store the parameters for the neural network is unknown. nnpfc_total_kilobyte_size can be a value between 0 and 2 32 Must be in the range -1.

[0198] nnpfc_reserved_zero_bit_b shall be equal to 0 in the bitstream. Decoders shall ignore NNPFC SEI messages where nnpfc_reserved_zero_bit_b is not 0.

[0199] nnpfc_payload_byte[i] may contain the i-th byte of the bitstream. The byte sequence nnpfc_payload_byte[i] for all present values ​​of i must be a complete bitstream in compliance with ISO / IEC 15938-17.

[0200] Neural-network post-filter activation (NNFPA)

[0201] The syntax structure for NNFPA is shown in Table 17.

[0202] [Table 17]

[0203] The NNPFA syntax structure in Table 17 may be signaled in the form of an SEI message. An SEI message that signals the NNPFA syntax structure in Table 17 may be referred to as an NNPFA SEI message.

[0204] The NNPFA SEI message can activate or deactivate the possible use of a target neural network post-processing filter identified by nnpfa_target_id for post-processing filtering of a picture set.

[0205] There may be multiple NNPFA SEI messages for the same picture if the post-processing filters are used for different purposes or filter different color components.

[0206] nnpfa_target_id can indicate the target neural network post-processing filter specified by one or more NNPFC SEI messages with the same nnpfc_id as nnfpa_target_id associated with the current picture.

[0207] The value of nnpfa_target_id is 0 to 2 32 Must be in the range 256 to 511 and 231 ~2 32 nnpfa_target_id values ​​in the range -2 may be reserved for future use. 31 ~2 32 NNPFA SEI messages with nnpfa_target_id in the range of -2 MUST be ignored.

[0208] An NNPFA SEI message with a particular value of nnpfa_target_id must not currently be present on the PU unless one or both of the following conditions are true:

[0209] - there is currently in the CLVS an NNPFC SEI message with the same nnpfc_id as the specific value of nnpfa_target_id present in the PU preceding the current PU in decoding order;

[0210] - There is currently an NNPFC SEI message with nnpfc_id equal to the specific value of nnpfa_target_id of the PU

[0211] When a PU includes both an NNPFC SEI message with a specific value of nnpfc_id and an NNPFA SEI message with an nnpfa_target_id that is the same as the specific value of nnpfc_id, the NNPFC SEI message must precede the NNPFA SEI message in decoding order.

[0212] A value of 1 for nnpfa_cancel_flag can indicate that the persistence of the target neural network post-processing filter set by any previous NNPFA SEI message with the same nnpfa_target_id as the current SEI message is canceled. That is, the target neural network post-processing filter will not be used any more unless it is activated by another NNPFA SEI message with the same nnpfa_target_id as the current SEI message and an nnpfa_cancel_flag equal to 0. A value of 0 for nnpfa_cancel_flag can indicate that nnpfa_persistence_flag will continue.

[0213] nnpfa_persistence_flag can indicate the persistence of the target neural network post-processing filter for the current layer. A value of 0 for nnpfa_persistence_flag can indicate that the target neural network post-processing filter can only be used for post-processing filtering for the current picture. A value of 1 for nnpfa_persistence_flag can indicate that the target neural network post-processing filter can be used for post-processing filtering for the current picture and all subsequent pictures in the current layer in output order until one or more of the following conditions is true:

[0214] - A new CLVS for the current layer is started

[0215] - Bitstream ends

[0216] - The picture in the current layer associated with the NNPFA SEI message that has the same nnpfa_target_id as the current SEI message and the same nnpfa_cancel_flag as 1 is output after the current picture in output order.

[0217] The target neural network post-processing filter is not applied to subsequent pictures in the current layer associated with an NNPFA SEI message that has the same nnpfa_target_id as the current SEI message and an nnpfa_cancel_flag equal to 1.

[0218] Post-filter hint

[0219] The syntax structure for post-filter hints is shown in Table 18.

[0220] [Table 18]

[0221] The post-filter hint syntax structure in Table 18 may be signaled in the form of an SEI message. An SEI message that signals the post-filter hint syntax structure in Table 18 can be referred to as a post-filter hint SEI message.

[0222] The postfilter hints SEI message can provide postfilter coefficients or correlation information for the design of a postfilter so that the decoded and output picture set can potentially be used for post-processing to obtain improved display quality.

[0223] A value of 1 for filter_hint_cancel_flag may indicate that the SEI message cancels the persistence of a previous post-filter hint SEI message in the output order that the SEI message applies to the current layer. A value of 0 for filter_hint_cancel_flag may indicate that post-filter hint information follows.

[0224] The filter_hint_persistence_flag may indicate the persistence of the post-filter hint SEI message for the current layer. A value of 0 for filter_hint_persistence_flag may indicate that the post-filter hint applies only to the currently decoded picture. A value of 1 for filter_hint_persistence_flag may indicate that the post-filter hint SEI message applies to the currently decoded picture and persists for all subsequent pictures in the current layer in output order until one or more of the following conditions are true:

[0225] - A new CLVS for the current layer is started

[0226] - Bitstream ends

[0227] - Pictures in the current layer of the AU associated with the post-filter hints SEI message are output after the current picture in output order.

[0228] filter_hint_size_y can indicate the vertical size of the filter coefficient or correlation array. The value of filter_hint_size_y must be in the range 1 to 15.

[0229] filter_hint_size_x can indicate the horizontal size of the filter coefficient or correlation array. The value of filter_hint_size_x must be in the range 1 to 15.

[0230] filter_hint_type may indicate the type of transmitted filter hint as shown in Table 19. The value of filter_hint_type shall be in the range of 0 to 2. filter_hint_type values ​​equal to 3 shall not exist in the bitstream. Decoders shall ignore post-filter hint SEI messages with filter_hint_type equal to 3.

[0231] [Table 19]

[0232] A value of 1 for filter_hint_chroma_coeff_present_flag can indicate that a filter coefficient for chroma is present. A value of 0 for filter_hint_chroma_coeff_present_flag can indicate that a filter coefficient for chroma is not present.

[0233] filter_hint_value[cIdx][cy][cx] can indicate the filter coefficients or cross-correlation matrix elements between the original signal and the decoded signal with 16-bit precision. The value of filter_hint_value[cIdx][cy][cx] is -2 31 +1~2 31 Must be in the range -1. cIdx indicates the associated color component, cy indicates the vertical counter and cx can indicate the horizontal counter. Depending on the value of filter_hint_type the following may apply:

[0234] - If the value of filter_hint_type is 0, the coefficients of a two-dimensional FIR (Finite Impulse Response) filter of size filter_hint_size_y*filter_hint_size_x may be transmitted.

[0235] - On the other hand, if the value of filter_hint_type is 1, the filter coefficients of two 1D FIR filters may be transmitted. In this case, the value of filter_hint_size_y must be 2. An index cy of 0 may indicate the filter coefficients of the horizontal filter, and an index cy of 1 may indicate the filter coefficients of the vertical filter. In the filtering process, the horizontal filter is applied first, and the result may be filtered by the vertical filter.

[0236] - Otherwise (if the value of filter_hint_type is 2), the transmitted hint can represent the cross-correlation matrix between the original signal s and the decoded signal s'.

[0237] The normalized cross-correlation matrix for the relevant color component identified by cIdx of size filter_hint_size_y * filter_hint_size_x may be defined as in Equation 4.

[0238]

Number

[0239] In Equation 4, s represents the sample array of the color component cIdx of the original picture, s' represents the corresponding array of the decoded picture, h represents the vertical height of the relevant color component, w represents the horizontal width of the relevant color component, bitDepth represents the bit depth of the color component. Also, OffsetY is the same as (filter_hint_size_y >> 1), OffsetX is the same as (filter_hint_size_x >> 1), the range of cy is 0 <= cy < filter_hint_size_y, and the range of cx is 0 <= cx < filter_hint_size_x.

[0240] The decoder can derive a Wiener post-filter from the cross-correlation matrix between the original signal and the decoded signal and the auto-cross-correlation matrix of the decoded signal.

[0241] Problems with the prior art

[0242] The NNPFC SEI message can provide a neural network post-filter, and the NNPFA SEI message can provide activation of the post-filter specified in the NNPFC SEI message to a picture set. The NNPFA SEI message can specify that the target post-filter is applied only to the current picture or to the current picture and subsequent pictures in output order until one of the following events occurs:

[0243] - New CLVS for the current layer begins

[0244] - Bitstream ends

[0245] - The picture in the current layer associated with the NNPFA SEI message that has the same nnpfa_target_id as the current SEI message and the same nnpfa_cancel_flag as 1 is output after the current picture in output order.

[0246] The above signaling design, which includes and persists NNPFC and NNPFA, can cause problems in situations such as those shown in Figure 5. In the example of Figure 5, a new NNPFC SEI message may be present in an AU that includes a picture unit (PU) with POC3. This can create a situation where it is unclear whether a neural network post-filter is applied to the picture unit (PU) that includes POC3. If so, the NNPFC is applied after the previous NNPFA activated the basic NNPFC SEI, but an update is provided. Here, the picture unit (PU) may be a NAL unit set that includes the VCL NAL units of a coded picture and their associated non-VCL VAL units.

[0247] Example

[0248] The present application proposes various embodiments to solve the problems described above. The embodiments proposed by the present application may be implemented individually or in combination of two or more.

[0249] In the following, NNPFC is the NNPFC syntax structure of Tables 1 and 2 and may be signaled in the form of an SEI message, where NNPFC may be the NNPFC SEI message. NNPFA is the NNPFA syntax structure of Table 17 and may be signaled in the form of an SEI message, where NNPFA may be the NNPFA SEI message. Post-filter hint is the post-filter hint syntax structure of Table 18 and may be signaled in the form of an SEI message, where post-filter hint may be the post-filter hint SEI message.

[0250] To solve the above problem, three options may be applied.

[0251] Option 1: Specify that the neural network postfilter applied to a picture is the filter in the NNPFC that immediately precedes this picture in the output order, as shown in Figure 6.

[0252] Option 2: Specify that the neural network post-filter applied to the picture is the filter in NNPFC that precedes NNPFA in the decoding order, as shown in Figure 7.

[0253] Option 3: As shown in Figure 8, when an NNPFC SEI message exists, if there is a previous NNPFA SEI message with the same ID, specify that the persistence of the message is canceled.

[0254] The embodiment proposed by the present application can be summarized as follows.

[0255] 1. When an NNPFC SEI message is associated with a current picture, a target NNPF that applies to or is associated with the current picture may be determined to be the filter described / transmitted in the last NNPFC SEI message that precedes the current picture in decoding order and has the same ID. That is, the target NNPF may be determined to be the NNPF of the last NNPFC SEI message among one or more NNPF SEI messages, which may precede the current picture in decoding order and have the same identifier as the NNPFC SEI message. The NNPFC SEI message may be referred to as a "candidate NNPFC SEI message."

[0256] 2. In addition, when determining an activated NNPFC SEI message, it may be further specified to ignore an NNPFC SEI message including a repetition of an NNPFC SEI message including a basic NNPF. That is, an NNPFC SEI message that precedes the current picture in decoding order and indicates the same NNPF as the NNPF (basic NNPF) of the first NNPFC SEI message among NNPFC SEI messages that have the same ID corresponds to a repetition of the first NNPFC SEI message, and therefore such an NNPFC SEI message may be ignored in determining the target NNPF. Here, one or more NNPFC SEI messages that precede the current picture in decoding order, have the same ID, and do not correspond to a repetition of an NNPFC SEI message including a basic NNPF may be referred to as "NNPFC SEI message candidate." For example, an NNPFC SEI message that corresponds to a repetition of an NNPFC SEI message including a basic NNPF may not be included in the NNPFC SEI message candidate.

[0257] Example 1

[0258] Example 1 is related to the above-mentioned Summary 1. According to Example 1, the NNPF of the last NNPFC SEI message among the NNPFC SEI messages that precede the current picture in decoding order and have the same identifier as the NNPFA SEI message may be determined as the target NNPF.

[0259] Specifically, for a particular picture in which an NNPF is activated, the target NNPF may be the NNPF identified by the last NNPFC SEI message having the same nnpfc_id as nnpfa_target_id, where the last NNPFC SEI message may precede the first VCL NAL unit of the current picture in decoding order.

[0260] Example 2

[0261] Example 2 is related to the above summaries 1 and 2. According to Example 2, among NNPFC SEI messages that precede the current picture in decoding order and have the same ID, an NNPFC SEI message that includes a basic NNPF corresponds to a repetition of the initial NNPFC SEI message, and therefore, such an NNPFC SEI message may be designated to be ignored.

[0262] Specifically, for a particular picture in which an NNPF is activated, the target NNPF may be the NNPF identified by the last NNPFC SEI message with nnpfc_id equal to nnpfa_target_id, where the last NNPFC SEI message may precede the first VCL NAL unit of the current picture in decoding order and may not be a repetition of the NNPFC SEI message containing the base NNPF.

[0263] Video encoding method and video decoding method

[0264] In the following, video encoding methods and video decoding methods according to various embodiments of the present application will be described.

[0265] FIG. 9 shows an example of a video encoding method, and FIG. 10 shows an example of a video decoding method.

[0266] Referring to FIG. 9, at least one neural network available as a post-processing filter may be determined, and information about the determined neural network may be encoded into at least one NNPFC SEI message (S910).

[0267] It may be determined whether a target NNPF applicable to the current picture is activated, and information about the determined target NNPF may be encoded in an NNPFA SEI message (S920). The step of determining whether a target NNPF is activated (S920) may include a step of determining the target NNPF, a step of determining whether to cancel persistence of the target NNPF, and a step of determining whether the target NNPF is persistent.

[0268] Post-filter coefficients or correlation information for designing a post-filter may be encoded into a post-filter hint SEI message (S930). The NNPFC SEI message, the NNPFA SEI message, and / or the post-filter hint SEI message may be included in the NNPF SEI message.

[0269] The target NNPF may be determined or specified according to various embodiments of the present application when an NNPF SEI message is applied to a current picture in the video decoding device 200. For example, when an NNPFC SEI message is associated with the current picture, the target NNPF applied to or associated with the current picture may be determined to be the filter described / conveyed in the last NNPFC SEI message that precedes the current picture in decoding order and has the same ID. In this case, when determining the activated NNPFC SEI message, it may be further specified to ignore NNPFC SEI messages that include repetitions of NNPFC SEI messages that include a basic NNPF.

[0270] 10, an SEI message for the NNPF to be applied to the current picture may be obtained from the bitstream. The SEI message for the NNPF may include an NNPFC SEI message, an NNPFA SEI message, and / or a post-filter hint SEI message.

[0271] When an SEI message for an NNPF is applied to a current picture, at least one neural network available as a post-processing filter may be determined based on at least one NNPFC SEI message included in the SEI message for the NNPF (S1010).

[0272] Based on at least one NNPFA SEI message obtained from the bitstream, it may be determined whether a target NNPF applicable to the current picture is activated (S1020). The step S1020 of determining whether a target NNPF is activated may include a step of determining a target NNPF, a step of determining whether to cancel persistence of the target NNPF, and a step of determining whether persistence of the target NNPF is present.

[0273] If the target NNPF is activated (determined or sustained), the target neural network post-processing filter may be applied to the current picture (S1030).

[0274] Various embodiments of the present application may be utilized to determine or specify a target NNPF. For example, if an NNPFC SEI message is associated with a current picture, the target NNPF that applies to or is associated with the current picture may be determined to be the filter described / transmitted in the last NNPFC SEI message that precedes the current picture in decoding order and has the same ID. In this case, when determining an activated NNPFC SEI message, it may be further specified to ignore NNPFC SEI messages that include repetitions of NNPFC SEI messages that include a basic NNPF.

[0275] FIG. 11 shows a video encoding / decoding method for an embodiment in which a target NNPF is determined based on whether it has the same identifier as the NNPFA SEI message and whether it precedes the current picture.

[0276] 11, it may be determined whether a specific NNPFC SEI message candidate, which is the ith of all n NNPFC SEI messages, precedes the current picture in decoding order and has the same identifier as the NNPFA SEI message (S1110). If the specific NNPFC SEI message candidate does not precede the current picture in decoding order or does not have the same identifier as the NNPFC SEI message, the specific NNPFC SEI message candidate may not be included in the NNPFC SEI message candidates (S1130). On the other hand, if the specific NNPFC SEI message candidate precedes the current picture in decoding order and has the same identifier as the NNPFA SEI message, the specific NNPFC SEI message candidate may be included in the NNPFC SEI message candidates (S1120).

[0277] It may be determined whether the determination of step S1110 has been performed for all NNPFC SEI messages (S1140). If the determination of step S1110 has not been performed for some NNPFC SEI messages, steps S1110 to S1130 may be performed for the next NNPFC SEI message (S1150). These steps may be repeated until the determination of all NNPFC SEI messages is completed, at which point NNPFC SEI message candidates may be constructed.

[0278] The target NNPF may be determined to be the NNPF of the last NNPFC SEI message candidate among the NNPFC SEI message candidates (S1160). That is, the target NNPF may be determined to be the NNPF of the last NNPFC SEI message that precedes the current picture in decoding order and has the same identifier as the NNPFC SEI message.

[0279] FIG. 12 shows a video encoding / decoding method for an embodiment in which the target NNPF is determined based not only on whether it has the same identifier as the NNPFA SEI message and whether it precedes the current picture, but also on whether it is a repetition of an NNPFC SEI message containing a basic NNPF.

[0280] 12, it may be determined whether an i-th candidate NNPFC SEI message among all n NNPFC SEI messages precedes the current picture in decoding order, has the same identifier as the NNPFC SEI message, and is a repetition of an NNPFC SEI message including a basic NNPF (S1210). Here, the NNPFC SEI message including the basic NNPF may be the initial NNPFC SEI message.

[0281] If a specific NNPFC SEI message candidate does not precede the current picture in decoding order, does not have the same identifier as the NNPFA SEI message, or is a repetition of an NNPFC SEI message including a basic NNPF, the specific NNPFC SEI message candidate may not be included in the NNPFC SEI message candidates (S1230). Conversely, if a specific NNPFC SEI message candidate precedes the current picture in decoding order, has the same identifier as the NNPFA SEI message, and is not a repetition of an NNPFC SEI message including a basic NNPF, the specific NNPFC SEI message candidate may be included in the NNPFC SEI message candidates (S1220).

[0282] It may be determined whether the determination of step S1210 has been performed for all NNPFC SEI messages (S1240). If the determination of step S1210 has not been performed for some NNPFC SEI messages, steps S1210 to S1230 may be performed for the next NNPFC SEI message (S1250). These steps may be repeated until the determination of all NNPFC SEI messages is completed, at which point NNPFC SEI message candidates may be constructed.

[0283] The target NNPF may be determined to be the NNPF of the last NNPFC SEI message candidate among the NNPFC SEI message candidates (S1260). That is, the target NNPF may be determined to be the NNPF of the last NNPFC SEI message that precedes the current picture in decoding order, has the same identifier as the NNPFA SEI message, and is not a repetition of the NNPFC SEI message that includes the basic NNPF.

[0284] FIG. 13 is a diagram illustrating a content streaming system to which an embodiment of the present disclosure can be applied.

[0285] As shown in FIG. 13, a content streaming system to which an embodiment of the present disclosure is applied may broadly include an encoding server, a streaming server, a web server, a media storage, a user device, and a multimedia input device.

[0286] The encoding server compresses content input from a multimedia input device such as a smartphone, camera, camcorder, etc. into digital data to generate a bitstream and transmits the bitstream to the streaming server. As another example, if a multimedia input device such as a smartphone, camera, camcorder, etc. directly generates a bitstream, the encoding server may be omitted.

[0287] The bitstream may be generated by a video encoding method and / or video encoding device to which an embodiment of the present disclosure is applied, and the streaming server may temporarily store the bitstream during the process of transmitting or receiving the bitstream.

[0288] The streaming server transmits multimedia data to a user device based on a user request via a web server, and the web server may act as an intermediary informing the user of available services. When a user requests a desired service from the web server, the web server transmits the request to the streaming server, which then transmits the multimedia data to the user. In this case, the content streaming system may include a separate control server, which may control commands and responses between devices in the content streaming system.

[0289] The streaming server can receive content from a media storage and / or encoding server. For example, when receiving content from the encoding server, the content can be received in real time. In this case, the streaming server can store the bitstream for a certain period of time to provide a smooth streaming service.

[0290] Examples of the user devices include mobile phones, smartphones, laptop computers, digital broadcasting terminals, personal digital assistants (PDAs), portable multimedia players (PMPs), navigation systems, slate PCs, tablet PCs, ultrabooks, wearable devices (e.g., smartwatches, smart glasses, and head-mounted displays (HMDs)), digital TVs, desktop computers, and digital signage.

[0291] Each server in the content streaming system may be operated as a distributed server, in which case data received by each server may be processed in a distributed manner.

[0292] The scope of the present disclosure includes software or machine-executable instructions (e.g., operating systems, applications, firmware, programs, etc.) that cause operations according to the methods of various embodiments to be performed on a device or computer, and non-transitory computer-readable media on which such software or instructions, etc., may be stored and executed on a device or computer.

[0293] [Industrial Applicability] The embodiments of the present disclosure can be used to encode / decode video.

[0294] [Claims at the time of international application] [Claim 1] A video decoding method performed by a video decoding device, comprising: obtaining a neural-network post-filter (NNPF) SEI (supplemental enhancement information) message; determining at least one neural network available as a post-processing filter based on the NNPF SEI message being applied to the current picture, based on at least one NNPFC (neural-network post-filter characteristics) SEI message included in the NNPF SEI message; determining whether a target neural-network post-filter activation (NNPFA) SEI message included in the NNPF SEI message is to be activated; The video decoding method, wherein the target neural network post-processing filter is determined as the neural network post-processing filter of the last NNPFC SEI message in decoding order among NNPFC SEI messages. [Claim 2] The video decoding method of claim 1 , wherein the NNPFC SEI message precedes the current picture in decoding order and has the same identifier as the NNPFA SEI message. [Claim 3] The video decoding method of claim 2 , wherein an NNPFC SEI message representing a neural network post-processing filter that is the same as the neural network post-processing filter of a first NNPFC SEI message in the NNPFC SEI message is not included in the NNPFC SEI message. [Claim 4] A video encoding method performed by a video encoding device, comprising: encoding at least one neural network usable as a post-processing filter into at least one NNPFC (neural-network post-filter characteristics) SEI (supplemental enhancement information) message; encoding in at least one NNPFA (neural-network post-filter activation) SEI message whether a target neural-network post-processing filter applicable to the current picture is activated; In the video decoding device, based on the fact that a neural-network post-filter (NNPF) SEI message is applied to a current picture, the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages; The video encoding method, wherein the NNPF SEI message includes the NNPFC SEI message and the NNPFA SEI message. [Claim 5] 5. The video encoding method of claim 4, wherein the NNPFC SEI message precedes the current picture in decoding order and has the same identifier as the NNPFA SEI message. [Claim 6] 6. The video encoding method of claim 5, wherein an NNPFC SEI message representing a neural network post-processing filter that is the same as the neural network post-processing filter of a first NNPFC SEI message in the NNPFC SEI message is not included in the NNPFC SEI message. [Claim 7] A computer-readable recording medium storing a bitstream generated by a video encoding method, The video encoding method includes: encoding at least one neural network usable as a post-processing filter into at least one NNPFC (neural-network post-filter characteristics) SEI (supplemental enhancement information) message; encoding in at least one NNPFA (neural-network post-filter activation) SEI message whether a target neural-network post-processing filter applicable to the current picture is activated; In the video decoding device, based on the fact that a neural-network post-filter (NNPF) SEI message is applied to a current picture, the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages; The NNPF SEI message includes the NNPFC SEI message and the NNPFA SEI message. [Claim 8] A computer-readable recording medium storing a bitstream generated by the video encoding method of claim 4.

Claims

1. A video decoding method performed by a video decoding device, comprising: obtaining a neural-network post-filter (NNPF) supplemental enhancement information (SEI) message; determining at least one neural network available as a post-processing filter based on at least one neural-network post-filter characteristics (NNPFC) SEI message included in the NNPF SEI message, based on the NNPF SEI message being applied to a current picture; determining whether a target neural network post-processing filter applicable to the current picture is activated based on at least one neural-network post-filter activation (NNPFA) SEI message included in the NNPF SEI message; The video decoding method, wherein the target neural network post-processing filter is determined to be a neural network post-processing filter of a last NNPFC SEI message in decoding order among NNPFC SEI messages.

2. The video decoding method of claim 1 , wherein the NNPFC SEI message precedes the current picture in decoding order and has the same identifier as the NNPFA SEI message.

3. The video decoding method of claim 2 , wherein an NNPFC SEI message representing a neural network post-processing filter that is the same as a neural network post-processing filter of a first NNPFC SEI message in the NNPFC SEI message is not included in the NNPFC SEI message.

4. A video encoding method performed by a video encoding device, comprising: encoding at least one neural network usable as a post-processing filter into at least one neural-network post-filter characteristics (NNPFC) supplemental enhancement information (SEI) message; encoding whether a target neural network post-processing filter applicable to the current picture is activated in at least one neural-network post-filter activation (NNPFA) SEI message; In the video decoding apparatus, based on the fact that a neural-network post-filter (NNPF) SEI message is applied to a current picture, the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages; The video encoding method, wherein the NNPF SEI message includes the NNPFC SEI message and the NNPFA SEI message.

5. The video encoding method of claim 4 , wherein the NNPFC SEI message precedes the current picture in decoding order and has the same identifier as the NNPFA SEI message.

6. The video encoding method of claim 5 , wherein an NNPFC SEI message representing a neural network post-processing filter that is the same as a neural network post-processing filter of a first NNPFC SEI message in the NNPFC SEI message is not included in the NNPFC SEI message.

7. A computer-readable recording medium storing a bitstream generated by a video encoding method, The video encoding method includes: encoding at least one neural network usable as a post-processing filter into at least one neural-network post-filter characteristics (NNPFC) supplemental enhancement information (SEI) message; encoding whether a target neural network post-processing filter applicable to the current picture is activated in at least one neural-network post-filter activation (NNPFA) SEI message; In the video decoding apparatus, based on the fact that a neural-network post-filter (NNPF) SEI message is applied to a current picture, the target neural-network post-processing filter is determined to be the neural-network post-processing filter of the last NNPFC SEI message in decoding order among the NNPFC SEI messages; The NNPF SEI message includes the NNPFC SEI message and the NNPFA SEI message.

8. A computer-readable recording medium storing a bitstream generated by the video encoding method of claim 4.

Images