Inter prediction method, encoder, decoder and computer readable storage medium

Abstract

The embodiment of the application discloses an inter-frame prediction method, an encoder, a decoder and a computer readable storage medium, comprising: parsing a code stream, determining a prediction decoding parameter of a current block; when the prediction decoding parameter indicates that an inter-frame geometric partition prediction mode (GPM) is used to determine the inter-frame prediction of the current block, determining GPM mode parameters and curve partition information of the current block; determining a first reference block and a second reference block based on the GPM mode parameters; determining a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve partition information; and determining a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

Description

[0001] This application requires a divisional application submitted to the China Patent Office on March 31, 2020, with application number 202010246244.3 and invention name "Inter-frame prediction method, encoder, decoder and computer-readable storage medium". Technical Field

[0002] The embodiments of the present application relate to video coding technology, and relate to but are not limited to an inter-frame prediction method, an encoder, a decoder, and a computer-readable storage medium. Background Art

[0003] In video encoding and decoding, in addition to intra-frame prediction, inter-frame prediction can also be used during the encoding and decoding of the current block. Inter-frame prediction can include motion estimation and motion compensation. For motion compensation, the inter-frame geometric partitioning prediction mode (GPM) can be used to divide the current block between frames into two non-rectangular partitions (or two blocks), perform predictions on each partition, and then perform weighted fusion to obtain the predicted value of the current block.

[0004] Currently, during GPM prediction, a straight line divides a rectangular block into two blocks. This geometric division can be done by dividing coding units or prediction units, collectively referred to as blocks. The resulting blocks are triangular or trapezoidal. The line dividing the rectangular block can have different angles and positions, resulting in a variety of triangular and trapezoidal combinations. This geometric division can align block edges closer to object edges or to the edges of two partitions with different motion vectors, thereby improving coding performance.

[0005] However, such (straight line) geometric division cannot completely fit the edge of the object, or the edge of two partitions with different motion vectors. Transition will be used for the edge area of ​​the geometric division. For some scenes, such as the existence of a large number of curves and arcs in natural objects, such straight line division is still too rigid, which affects the performance of encoding and decoding. Summary of the Invention

[0006] The embodiments of the present application provide an inter-frame prediction method, an encoder, a decoder, and a computer-readable storage medium, which can improve the accuracy of encoding and decoding.

[0007] The technical solution of this application is achieved as follows:

[0008] An embodiment of the present application provides an inter-frame prediction method, which is applied in a decoder and includes:

[0009] Parse the code stream and determine the prediction decoding parameters of the current block;

[0010] When the prediction decoding parameter indicates that the inter-frame geometric partitioning prediction mode GPM is used to determine the inter-frame prediction of the current block, determining the GPM mode parameters and curve partitioning information of the current block;

[0011] determining a first reference block and a second reference block based on the GPM mode parameters;

[0012] Determining, based on the GPM mode parameters and the curve division information, a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block;

[0013] A prediction value of the current block is determined based on the first prediction parameter and the second prediction parameter.

[0014] The embodiment of the present application further provides an inter-frame prediction method, which is applied to an encoder and includes:

[0015] Determining prediction mode parameters for the current block;

[0016] When the prediction mode parameter indicates that the inter-frame geometric partitioning prediction mode GPM is used to determine the inter-frame prediction of the current block, determining the GPM mode parameter and curve partitioning information of the current block;

[0017] determining a first reference block and a second reference block based on the GPM mode parameters;

[0018] Determining, based on the GPM mode parameters and the curve division information, a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block;

[0019] A prediction value of the current block is determined based on the first prediction parameter and the second prediction parameter.

[0020] An embodiment of the present application provides a decoder, including:

[0021] A decoding unit, used to parse the code stream and determine the prediction decoding parameters of the current block;

[0022] a first determining unit configured to, when the prediction decoding parameter indicates that an inter-frame geometric partitioning prediction mode GPM is used to determine inter-frame prediction of the current block, determine GPM mode parameters and curve partitioning information of the current block; determine a first reference block and a second reference block based on the GPM mode parameters; and determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve partitioning information;

[0023] The first prediction unit is configured to determine a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

[0024] An embodiment of the present application provides an encoder, including:

[0025] a second determining unit configured to determine a prediction mode parameter of the current block; when the prediction mode parameter indicates that an inter-frame geometric partitioning prediction mode (GPM) is used to determine inter-frame prediction of the current block, determine the GPM mode parameter and curve partitioning information of the current block; determine a first reference block and a second reference block based on the GPM mode parameter; and determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameter and the curve partitioning information;

[0026] The second prediction unit is configured to determine a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

[0027] An embodiment of the present application also provides a decoder, comprising: a first processor and a first memory storing executable instructions of the first processor, wherein the first memory relies on the first processor to perform operations through a communication bus, and when the executable instructions are executed by the first processor, the inter-frame prediction method described in any one of the embodiments of the present application is executed.

[0028] An embodiment of the present application also provides an encoder, comprising: a second processor and a second memory storing executable instructions of the second processor, wherein the second memory relies on the second processor to perform operations through a communication bus, and when the executable instructions are executed by the second processor, the inter-frame prediction method described in any one of the embodiments of the present application is executed.

[0029] An embodiment of the present application provides a computer-readable storage medium storing executable instructions. When the executable instructions are executed by one or more first processors, the first processors execute the inter-frame prediction method on the decoder side; or, when the executable instructions are executed by one or more second processors, the second processors execute the inter-frame prediction method on the encoder side.

[0030] The embodiments of the present application provide an inter-frame prediction method, encoder, decoder, and computer-readable storage medium, including: parsing a bitstream to determine prediction decoding parameters for a current block; when the prediction decoding parameters indicate the use of an inter-frame geometric partitioning prediction mode (GPM) to determine the inter-frame prediction of the current block, determining the GPM mode parameters and curve partitioning information for the current block; determining a first reference block and a second reference block based on the GPM mode parameters; determining a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve partitioning information; and determining a prediction value for the current block based on the first prediction parameter and the second prediction parameter. In other words, in the embodiments of the present application, by predicting the current block using curve partitioning in the GPM mode, the partitioned boundary is more closely aligned with the edge of an object or the edge of two parts with different motion vectors, thereby improving the accuracy of encoding and decoding. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic block diagram of a video encoding system according to an embodiment of the present application;

[0032] Figure 2 A schematic block diagram of a video decoding system according to an embodiment of the present invention;

[0033] Figure 3 A schematic diagram of an implementation flow of an inter-frame prediction method provided in an embodiment of the present application;

[0034] Figures 4a-4i A schematic diagram of an exemplary geometric partitioning method used by VVC provided in an embodiment of the present application;

[0035] Figure 5 A schematic diagram of the division angles of an exemplary VVC geometric division provided in an embodiment of the present application;

[0036] Figure 6 A schematic diagram of an exemplary transformation from a straight dividing line to a curved dividing line provided in an embodiment of the present application;

[0037] Figure 7 Schematic diagram of the implementation process of another inter-frame prediction method provided in the embodiment of the present application Figure 1 ;

[0038] Figure 8 Schematic diagram of the implementation process of another inter-frame prediction method provided in the embodiment of the present application Figure 2 ;

[0039] Figure 9 A schematic diagram of the structure of a decoder provided in an embodiment of the present application Figure 1 ;

[0040] Figure 10A structure diagram of a decoder provided for an embodiment of the present application Figure 2 ;

[0041] Figure 11 A structure diagram of an encoder provided for an embodiment of the present application Figure 1 ;

[0042] Figure 12 A structure diagram of an encoder provided for an embodiment of the present application Figure 2 . DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

[0044] The present application provides a video coding system, as shown in Figure 1 The video coding system 11 comprises:

[0045] Transformation unit 111, quantization unit 112, mode selection and coding control logic unit 113, intra-frame prediction unit 114, inter-frame prediction unit 115 (including motion compensation and motion estimation), inverse quantization unit 116, inverse transformation unit 117, loop filtering unit 118, coding unit 119 and decoded image buffer unit 110; for the input original video signal, through the coding tree block (Coding Tree A video reconstructed block can be obtained by dividing the video reconstructed block into a plurality of frames (CTUs). The coding mode is determined by the mode selection and coding control logic unit 113. Then, the residual pixel information obtained after intra-frame or inter-frame prediction is transformed by the transform unit 111 and the quantization unit 112 to transform the video reconstructed block, including transforming the residual information from the pixel domain to the transform domain and quantizing the obtained transform coefficients to further reduce the bit rate. The intra-frame prediction unit 114 is used to perform intra-frame prediction on the video reconstructed block. The intra-frame prediction unit 114 is used to determine the optimal intra-frame prediction mode (i.e., target prediction mode) of the video reconstructed block. The inter-frame prediction unit 115 is used to perform inter-frame prediction coding of the received video reconstructed block relative to one or more blocks in one or more reference frames to provide temporal prediction information. The motion estimation is used to generate motion information. The motion vector is used to estimate the motion of the video reconstructed block. Motion compensation is then performed based on the motion vector determined by motion estimation. After determining the inter-frame prediction mode, the inter-frame prediction unit 115 is further configured to provide the selected inter-frame prediction data to the encoding unit 119, and also to send the calculated motion vector data to the encoding unit 119. In addition, the inverse quantization unit 116 and the inverse transform unit 117 are configured to reconstruct the video reconstructed block, reconstructing a residual block in the pixel domain. This reconstructed residual block is passed through the loop filter unit 118 to remove blocking artifacts. The reconstructed residual block is then added to a predictive block in the frame of the decoded image buffer unit 110 to generate a reconstructed video reconstructed block. The encoding unit 119 is configured to encode various coding parameters and quantized transform coefficients. The decoded image buffer unit 110 is configured to store the reconstructed video reconstructed block for prediction reference. As video encoding progresses, new reconstructed video reconstructed blocks are continuously generated and stored in the decoded image buffer unit 110.

[0046] The embodiment of the present application provides a video decoding system, Figure 2 This is a schematic diagram of the composition structure of the video encoding system according to an embodiment of the present application. Figure 2 As shown, the video encoding system 12 includes:

[0047] The decoded unit 121, the inverse transformation unit 127, the inverse quantization unit 122, the intra prediction unit 123, the motion compensation unit 124, the loop filter unit 125 and the decoded image buffer unit 126; the video signal input is processed by the video encoding system 11 to output the code stream of the video signal; the code stream is input into the video decoding system 12, first passes through the decoded unit 121, and is used to obtain the decoded transform coefficient; the transform coefficient is processed by the inverse transformation unit 127 and the inverse quantization unit 122, so as to generate the residual block in the pixel domain; the intra prediction unit 123 can be used to generate the prediction data of the current video decoding block based on the determined intra prediction direction and the data from the previous decoded block of the current frame or picture; the motion compensation unit 124 is used to determine the prediction information for the video decoding block by parsing the motion vector and other associated syntax elements, and use the prediction information to generate the predictive block of the video decoding block being decoded; the decoded video block is formed by summing the residual block from the inverse transformation unit 127 and the inverse quantization unit 122 and the corresponding predictive block generated by the intra prediction unit 123 or the motion compensation unit 124; the decoded video signal passes through the loop filter unit 125 to remove the blocking artifact, and the video quality can be improved; then the decoded video block is stored in the decoded image buffer unit 126, the decoded image buffer unit 126 stores the reference image for subsequent intra prediction or motion compensation, and also is used for the output of the video signal to obtain the recovered original video signal.

[0048] The inter prediction method provided by the embodiment of the application mainly acts on the inter prediction unit 215 of the video encoding system 11 and the inter prediction unit, i.e. the motion compensation unit 124, of the video decoding system 12; that is to say, if a better prediction effect can be obtained by the inter prediction method provided by the embodiment of the application in the video encoding system 11, then correspondingly, the video decoding recovery quality can also be improved at the decoding end.

[0049] Based on this, the technical solution of the application is further described in detail in combination with the drawings and embodiments. Before the detailed description, it should be noted that the "first", "second", "third" and the like mentioned throughout the specification are only for distinguishing different features, and do not have the functions of limiting priority, sequence, size relationship and the like.

[0050] The embodiment of the application provides an inter prediction method, which is applied to a video decoding device, i.e. a decoder. The function realized by the method can be realized by calling program code by a first processor in the video decoding device, and of course the program code can be saved in a first memory, so the video decoding device at least includes the first processor and the first memory.

[0051] Figure 3This is a schematic diagram of an implementation flow of an inter-frame prediction method according to an embodiment of the present application. Figure 3 As shown, the method includes:

[0052] S101: parse the code stream to determine the prediction decoding parameters of the current block.

[0053] S102 : When the prediction decoding parameter indicates using GPM to determine inter-frame prediction of the current block, determine GPM mode parameters and curve division information of the current block.

[0054] S103. Determine a first reference block and a second reference block based on GPM mode parameters.

[0055] S104 : Determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve division information.

[0056] S105 . Determine a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

[0057] In an embodiment of the present application, during the decoding process, the decoder can parse out information such as the encoding method used by the encoder during encoding from the bit stream, that is, the predicted decoding parameters for the current block. Based on the diversity of current encoding methods, the prediction modes decoded by the decoder may be diverse. The inter-frame prediction method provided in the embodiment of the present application is a scenario that can only be implemented when the GPM prediction mode is used for encoding.

[0058] When the decoder parses the bitstream to obtain the prediction decoding parameters indicating the use of the inter-frame geometric partitioning prediction mode GPM to determine the inter-frame prediction of the current block, the inter-frame prediction method provided in the embodiment of the present application is adopted to parse the GPM mode parameters and curve partitioning information of the current block used in encoding from the bitstream.

[0059] The GPM mode parameters include a GPM division mode and motion related information. The GPM division mode can be 64.

[0060] In an embodiment of the present application, the decoder decodes the prediction mode of the current block. If the current block uses GPM, the GPM needs to decode the GPM partition mode. If the GPM uses a mask method, the MV information used by the two partitions of the current block or other GPM-related information is decoded.

[0061] For the curve division mode, the decoder can directly determine it by decoding the curve division usage flag in the bitstream. For example, if the curve division usage flag is 1 (indicating that it is used), it indicates that the current block uses curve division, and the translation amplitude and / or curve shape index transmitted in the bitstream can be decoded. The decoding order of the translation amplitude and curve shape index can be reversed, which is not limited by the embodiment of the present application. If the curve division usage flag is 0 (indicating that it is not used), the decoding of the curve division information is stopped.

[0062] In an embodiment of the present application, for the curve division mode, the decoder can indirectly determine it by decoding the translation amplitude in the code stream. If the translation amplitude is not 0 (indicating that the curve division mode is used), it means that curve division is used and the curve shape index continues to be decoded. If the translation amplitude is 0 (indicating that the curve division mode is not used), the decoding of the curve division information is stopped. Among them, there may be multiple combinations of the order of decoding the curve division information, the division mode of the GPM, the information of the MV used for the two division parts of the current block, or other information related to the GPM, which is not limited in the embodiment of the present application. The decoding method of the division mode of the GPM, the information of the MV used for the two division parts of the current block, or other information related to the GPM is not limited in this application.

[0063] It should be noted that, in an embodiment of the present application, a video image can be divided into multiple image blocks, and each image block currently to be decoded can be called a decoding block, wherein each decoding block may include a first image component, a second image component, and a third image component; and the current block is a decoding block in the video image that is currently to be predicted for the first image component, the second image component, or the third image component.

[0064] Here, assuming that the current block performs a first image component prediction, and the first image component is a luminance component, that is, the image component to be predicted is a luminance component, then the current block can also be called a luminance block; or, assuming that the current block performs a second image component prediction, and the second image component is a chrominance component, that is, the image component to be predicted is a chrominance component, then the current block can also be called a chrominance block.

[0065] It should also be noted that the prediction mode parameter indicates the coding mode of the current block and parameters related to the mode.

[0066] In an embodiment of the present application, the division mode of the GPM may include: angle index information, and may also include: step index information and target division mode, etc., which is not limited in the embodiment of the present application.

[0067] It should be noted that the current GPM geometric division can divide any shape that can be calculated by a geometric function. Considering the application scenario and implementation complexity, the existing geometric division method divides a rectangular block into two blocks by a straight line. Here, the geometric division may be the division of a coding unit (CU) or the division of a prediction unit (PU), which is collectively referred to as a current block. The divided blocks are triangular, rectangular or trapezoidal. The straight line that divides the rectangular block into two blocks can have different angles and different positions, so that a variety of combinations of triangles, rectangles and trapezoids can be divided. Usually, a transition is used for the edge region of the geometric division, such as using a weighted average method to use a gradual coefficient transition at the division edge, so that the division edge is not too harsh. The current geometric division method used by VVC. It should be noted that in the embodiments of the present application, GPM is used to divide the edge part of the object in the image into two non-rectangular sub-blocks for prediction and then weighted fusion. For example, the non-rectangular form can be as shown in Figures 4a-4i

[0068] As shown in Figure 5 , the geometric division of VVC divides the angle of a circle into 32 parts, and the offset of the straight line relative to the center of the block can have 4 different values, one of which is 0, that is, the straight line passes through the center of the block. The current VVC uses part of the 32 angles, and the offset method is different for different angles, and some of the existing division modes such as binary tree and ternary tree are excluded.

[0069] In the embodiments of the present application, after the decoder obtains the GPM mode parameters and the curve division information of the current block, the decoder can determine the first reference block and the second reference block corresponding to different partitions based on the division mode of the GPM in the GPM mode parameters and the motion-related information, and then determine the first prediction parameter corresponding to the first reference block and the second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve division information.

[0070] In the embodiments of the present application, the first prediction parameter includes a first weight value or first boundary position information, and the second prediction parameter includes a second weight value or second boundary position information.

[0071] It should be noted that the first weight value and the second weight value are the weight values corresponding to the pixel points when the GPM division is performed using the mask method, and the first boundary position information and the second boundary position information are the boundary information of the first reference block and the second reference block.

[0072] ​In the embodiment of the present application, the decoder can determine the first prediction parameter corresponding to the first reference block and the second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve partitioning information in two ways based on the difference in the prediction parameters, as follows:

[0073] Method 1: The first prediction parameter includes: a first weight value; the second prediction parameter includes: a second weight value; the decoder determines a preset weight distribution corresponding to the current GPM mode based on the GPM mode parameter; determines the size information of the current block; updates the preset weight distribution based on the size information, the GPM mode parameter and the curve division information to obtain a curve weight distribution; obtains the first weight value and the second weight value based on the curve weight distribution.

[0074] It should be noted that the curve division information includes: translation amplitude and curve shape index; the decoder determines the offset direction according to the GPM mode parameters; determines the offset amount according to the size information, translation amplitude and curve shape index; according to the offset direction, the preset weight of each pixel position in the preset weight distribution is offset by the offset amount to obtain the updated curve weight of each pixel position, thereby obtaining the curve weight distribution.

[0075] Method 2: The first prediction parameter includes: first boundary position information; the second prediction parameter includes: second boundary position information; the decoder obtains the size information of the current block; based on the GPM mode parameter and the size information, the division boundary position information is determined; according to the GPM mode parameter, the size information and the curve division information, the division boundary position information is offset to determine the curve division boundary position information; based on the curve division boundary position information, the first boundary position information and the second boundary position information are determined.

[0076] It should be noted that the curve division information includes: translation amplitude and curve shape index; the decoder determines the offset direction according to the GPM mode parameters; determines the offset amount according to the size information, translation amplitude and curve shape index; according to the offset direction, the position of each pixel point of the division boundary position information is offset by the offset amount to obtain the curve division boundary position information.

[0077] In Method 1, the preset weight distribution refers to the original mask corresponding to the GPM division mode using straight-line division. The curved weight distribution is the weight distribution information corresponding to the curved division mode, which is updated by performing a translation calculation on the preset weight distribution. The decoder primarily uses the size information, GPM mode parameters, and curved division information to update the preset weight distribution to obtain the curved weight distribution. From the curved weight distribution, the first weight value corresponding to the first reference block and the second weight value corresponding to the second reference block are obtained for each pixel of the current block.

[0078] In some embodiments of the present application, the decoder determines the offset direction based on the angle information in the GPM mode parameters, and determines the required offset amount based on the size information, the translation amplitude, and the curve shape index.

[0079] In method 2, the curve division method adopted by the decoder for the current block can use the pixel value of the first reference block as the first prediction value for part of the first reference block, and use the pixel value of the second reference block as the second prediction value for part of the second reference block. However, at the boundary position, the pixel points on the respective boundary information of the first reference block and the second reference block are offset to obtain the first boundary position information and the second boundary position information, and then the pixel points on the boundary are predicted based on the first boundary position information and the second boundary position information.

[0080] It should be noted that the GPM mode parameters are used to divide the current block. Based on the size information of the current block, the division boundary position information of the current block can be obtained; at this time, the decoder determines the offset direction according to the GPM mode parameters; determines the offset amount according to the size information, translation amplitude and curve shape index; according to the offset direction, the position of each pixel point of the division boundary position information is offset by the offset amount to obtain the curve division boundary position information.

[0081] In an embodiment of the present application, the decoder decodes the block partition information. If the current block uses GPM, the GPM needs to decode the GPM partition mode.

[0082] For example, if GPM uses the method of dividing CU or PU, the division boundary is determined according to the GPM mode information and curve division information. If the division boundary determined according to the GPM mode information is boundary, (x n ,y n ) is a point on the boundary. The decoder determines the boundary of the division as boundary' according to the GPM mode information and curve division information, (x' n ,y′ n ) is a point on boundary'.

[0083] In response to the above implementation, in an embodiment of the present application, the decoder determines the offset direction based on the GPM mode parameters as follows: when the angle used by the GPM represented by the GPM mode parameters meets the horizontal judgment condition, the offset direction is determined to be the horizontal direction; when the angle used by the GPM represented by the GPM mode parameters meets the vertical judgment condition, the offset direction is determined to be the vertical direction.

[0084] In some embodiments of the present application, the decoder determines the offset direction according to the angle information in the GPM mode parameters.

[0085] In some embodiments of the present application, the vertical judgment condition is that the angle is within the horizontal angle range; the horizontal judgment condition is that the angle is within the vertical angle range, or other conditions other than the horizontal judgment condition are vertical directions. The embodiments of the present application do not limit the judgment conditions, as long as the method can distinguish the moving direction, it is acceptable.

[0086] For example, when the angle information in the GPM partition parameter, i.e., angle is 4 to 12 (excluding 12) or angle is 20 to 28 (excluding 28), the offset direction is vertical; otherwise, the offset direction is horizontal. This is not limited in this embodiment of the present application. That is, the vertical determination condition can be If((angleIdx>=4&&angleIdx<12)||(angleIdx>=20&&angleIdx<28)).

[0087] In the following, for different offset modes, the decoder determines the description of the offset based on the size information, translation amplitude, and curve shape index.

[0088] In this embodiment of the present application, the size information includes the length and width of the current block.

[0089] When the offset direction is horizontal, the first pixel position in the preset block in the non-offset direction at each pixel position is determined according to each pixel position, the length, and the number of preset offset points, wherein the non-offset direction is perpendicular to the offset direction; first curve information corresponding to the first pixel position is determined according to the curve shape index; a first movement ratio is determined based on the width and the number of preset offset points; an offset amount is determined according to the first curve information, the translation amplitude, and the first movement ratio, wherein the offset amount does not exceed the width of the current block.

[0090] It should be noted that the number of preset offset points is the number of points to be offset corresponding to different curve division modes. In the embodiment of the present application, the number of preset offset points can be 64. The embodiment of the present application does not limit this. The more preset offset points there are, the better the curve division.

[0091] In the embodiment of the present application, when the offset direction is the horizontal direction, the non-offset direction is the vertical direction; when the offset direction is the vertical direction, the non-offset direction is the horizontal direction.

[0092] In the embodiment of the present application, the decoder multiplies the first curve information with the translation amplitude and the first movement ratio to obtain an offset.

[0093] In some embodiments of the present application, the decoder determines the first pixel position in the preset block in the non-offset direction at each pixel position according to the length and the number of preset offset points based on the following steps: determining a first scaling ratio according to the length and the number of preset offset points; determining a first sub-offset amount according to the length and the number of preset offset points; and determining the first pixel position in the preset block in the non-offset direction at each pixel position based on the first scaling ratio and the first sub-offset amount.

[0094] In the embodiments of the present application, the decoder obtains the first scaling ratio by multiplying the ordinate of each pixel position by the proportion of the number of preset offset points and the length, or obtains the first scaling ratio by using the logarithm with base 2 of the length and the number of preset offset points and a shift; the decoder then obtains the first sub-offset amount according to the number of preset offset points and (2*length); finally, the decoder adds the first scaling ratio and the first sub-offset amount to obtain the first pixel position. The number of preset offset points can be 64, 32, 16, 8, etc., and the embodiments of the present application do not make any limitation.

[0095] For example, assuming that the decoder obtains the mask P' (curve weight distribution) from the mask P (preset weight distribution) according to the translation amplitude used by the curve division and the curve shape and the mask P, and assuming that the translation amplitude is curveAmplitude, the curve shape index used by the translation is curveIdx, there are NUM_CURVE curve shapes in total, the LUT (curve information) is curveFactor[NUM_CURVE][N], and N is equal to, for example, the width of the current block is width, the length of the current block is height, and the angle used by the GPM is angle.

[0096] If the current block uses the horizontal direction, the P' of each pixel point (x, y) is as shown in formula (1) as follows:

[0097] P'[x][y]= P[x-offsetX][y] (1)

[0098] wherein, offsetX is the offset amount. The offset amount is obtained according to formula (2) as follows:

[0099] offsetX= curveFactor[curveIdx][y N ]* curveAmplitude *Scale1 (2)

[0100] Here, curveFactor[curveIdx][y N ] is the first curve information, curveAmplitude is the translation amplitude, and Scale1 is the first moving ratio.

[0101] Here, the first moving ratio is obtained by formula (3), y N The following is obtained for formula (4):

[0102] Scale1=width / N (3)

[0103] y N =y*N / height+offset1 (4)

[0104] Wherein, offset1 is the first sub-offset, which is determined by formula (5) as follows:

[0105] Offset1=N / (2*height) (5)

[0106] It should be noted that x-offsetX will not exceed the normal range of the current block, such as 0 to width-1.

[0107] In some embodiments of the present application, when the offset direction is a vertical direction, the decoder determines the second pixel position in the preset block in the non-offset direction at each pixel position based on each pixel position, the width, and the number of preset offset points, wherein the non-offset direction is perpendicular to the offset direction; determines the second curve information corresponding to the second pixel position based on the curve shape index; determines the second movement ratio based on the length and the number of preset offset points; determines the offset amount based on the second curve information, the translation amplitude, and the second movement ratio, wherein the offset amount does not exceed the length of the current block.

[0108] In the embodiment of the present application, the decoder multiplies the second curve information with the translation amplitude and the second movement ratio to obtain an offset.

[0109] In some embodiments of the present application, the decoder determines the second pixel position located in the preset block in the non-offset direction at each pixel position based on each pixel position, the width, and the number of preset offset points as follows: the decoder determines a second scaling ratio based on each pixel position, the width, and the number of preset offset points; determines a second sub-offset based on the width and the number of preset offset points; and determines the second pixel position located in the preset block in the non-offset direction at each pixel position based on the second scaling ratio and the second sub-offset.

[0110] In an embodiment of the present application, the decoder multiplies the horizontal coordinate of each pixel position by the ratio of the number of preset offset points to the width to obtain a second scaling ratio, or the decoder uses the logarithm with base 2 of the width and the number of preset offset points and shifts to obtain the second scaling ratio; the decoder then obtains a second sub-offset based on the number of preset offset points and (2*width); finally, the decoder adds the second scaling ratio and the second sub-offset to obtain the second pixel position.

[0111] For example, assume that the decoder derives mask P' (curve weight distribution) from the translation amplitude and curve shape used for curve segmentation and mask P (preset weight distribution). Let the translation amplitude be curveAmplitude, the index of the curve shape used for translation be curveIdx, there are NUM_CURVE curve shapes in total, and the LUT (curve information) be curveFactor[NUM_CURVE][N], where N is constant. Let the width of the current block be width, the length be height, and the angle used by GPM be angle.

[0112] If the current block uses the vertical direction, P' of each pixel (x, y) is as shown in formula (6):

[0113] P'[x][y]= P[x][y-offsetY] (6)

[0114] Where offsetY is the offset. The offset is obtained by formula (7), as follows:

[0115] offsetX= curveFactor[curveIdx][x N ]* curveAmplitude *Scale2 (7)

[0116] Here, curveFactor[curveIdx][x N ] is the second curve information, curveAmplitude is the translation amplitude, and Scale2 is the second movement scale.

[0117] Here, the second moving ratio is obtained by formula (8), x N The following is obtained for formula (9):

[0118] Scale2= height / N (8)

[0119] x N =x*N / width+offset2 (9)

[0120] Wherein, offset2 is the second sub-offset, which is determined by formula (10) as follows:

[0121] offset2=N / (2*width) (10)

[0122] It should be noted that y-offsetY will not exceed the normal range of the current block, such as 0 to height-1.

[0123] Based on the above two offset directions, another way to express the offset process is:

[0124] If((angleIdx>=4&&angleIdx<12)||(angleIdx>=20&&angleIdx<28))

[0125] offsetCX=0

[0126] offsetCY=curveFactor[curveIdx][xL*N / nW+N / 2nW]*curveAmplitude*nH / N

[0127] Otherwise(shiftHor is equal to 1), the following applies:

[0128] offsetCX=curveFactor[curveIdx][yL*N / nH+N / 2nH]*curveAmplitude*nW / N

[0129] offsetCY=0

[0130] Where nW is the width, nH is the length, offsetCX is offsetX, offsetCY is offsetY, and angleIdx is the angle index.

[0131] In some embodiments of the present application, the decoder may further first determine the position (x', y') of each pixel point (x, y) of the current block on the preset weight distribution based on the translation amplitude and curve shape used for curve division, and then calculate predSamples[x][y] for each pixel point (x, y) based on the preset weight distribution, the offset direction and the offset amount to temporarily calculate the required P[x'][y'].

[0132] For example, if the current block uses the horizontal direction, x'=x-offsetX, y'=y

[0133] If the current block uses the vertical direction, x'=x, y'=y-offsetY

[0134] Where offsetX and offsetY are obtained in the same way as above.

[0135] It should be noted that in this embodiment of the present application, the decoder may first calculate the mask P, and then calculate predSamples[x][y] for each point (x, y). It is also possible to temporarily calculate the required P[x][y] when calculating predSamples[x][y] for each point (x, y). In this case, the curve partitioning method is used to temporarily calculate the required P[x'][y'] when calculating predSamples[x][y] for each point (x, y). This embodiment of the present application does not limit this, and the final purpose is the same.

[0136] After the decoder obtains the first weight value and the second weight value using method 1, the decoder can determine the prediction value of the current block based on the first weight value, the first pixel value of the first reference block, the second weight value and the second pixel value of the second reference block.

[0137] In an embodiment of the present application, the decoder performs a weighted sum operation on the first weight value, the first pixel value of the first reference block, the second weight value, and the second pixel value of the second reference block to obtain a prediction value of the current block.

[0138] For example, let the pixel value at position (x, y) of the first reference block be predSamplesLA[x][y], and the pixel value at position (x, y) of the second reference block be predSamplesLB[x][y]. The decoder obtains the mask P according to the GPM division mode. Each pixel position (x, y) on the mask has a value P[x][y] indicating the weight of the first reference block (x, y). M is the sum of the weights of the corresponding positions of the two reference blocks. The weight of the second reference block (x, y) is MP[x][y]. The decoder obtains the mask P' based on the translation amplitude and curve shape used in the curve division and the mask P.

[0139] That is, the predicted value of a pixel point (x, y) in the current block is shown in formula (11):

[0140] predSamples[x][y]=(P'[x][y]*predSamplesLA[x][y]+(M-P'[x][y])*predSamplesLB

[0141] [x][y]) / M(11)

[0142] Among them, predSamples[x][y] is the predicted value of a pixel point (x, y) in the current block.

[0143] It should be noted that M can be a power of 2, such as 4, 8, 16, etc., so that the division in the equation can be written as a right shift. If predSamples[x][y] exceeds the normal value range, it needs to be processed to be within the range, which is not restricted by this application.

[0144] After the decoder obtains the first boundary position information and the second boundary position information using the second method, the decoder determines the prediction value of the current block based on the first prediction parameter and the second prediction parameter as follows: the decoder divides the current block according to the first boundary position information and the second boundary position information to obtain a first block and a second block; predicts the first block based on the first pixel value of the first reference block, and predicts the second block based on the second pixel value of the second reference block, thereby determining the prediction value of the current block.

[0145] It is understandable that the decoder predicts the current block by using curve partitioning in the GPM mode, so that the partitioned boundary is more closely aligned with the edge of the object or the edge of two parts with different motion vectors, thereby improving the accuracy of encoding and decoding.

[0146] It should be noted that after obtaining the prediction block of the current block, the decoder decodes the residual information of the current block, including the decoding coefficients, dequantizes and inversely transforms to obtain the residual block, and adds the residual block to the prediction block to obtain the decoded block of the current block.

[0147] The inter-frame prediction provided in the embodiment of the present application adopts a curve division method.

[0148] In an embodiment of the present application, the curve division information is a division method in curve division; curve division is to translate the pixel points on the direct division line in the preset straight line geometric division to obtain a continuous curve division line to realize geometric division.

[0149] In some embodiments of the present application, each pixel point of a straight line dividing line is continuously translated according to different moving distances and the same moving direction to obtain a curved line dividing line; or, each pixel point of a straight line dividing line is continuously translated according to different moving vectors to obtain a curved line dividing line.

[0150] That is to say, if the points on a straight line segment (the straight lines mentioned here are all line segments) are translated in the same direction, the translation distance of each point is not the same, the distances of adjacent points are not much different, and the continuity of the line is maintained, a curved segment can be obtained. Or if the points on a straight line segment are translated, the translation vector of each point is not the same, the vectors of adjacent points are not much different, and the continuity of the line is maintained, a curved segment can also be obtained. For video encoding and decoding, the so-called straight line is composed of points whose pixel positions are on the same line or approximately on the same line. If these points are translated in the same direction, the translation distance of each point is not the same, and the distances of adjacent points meet certain requirements, a curved segment can be obtained. If these points are translated, the translation vector of each point is not the same, and the distances of adjacent points meet certain requirements, a curved segment can be obtained.

[0151] For example, Figure 6 The following diagram shows the transformation from a straight line to a curve. By translating the points on the boundary line of two blocks in the geometric partitioning as shown by the arrows in the figure above, the boundary line can be transformed into a curved segment, and the straight partitioning can be transformed into a curved partitioning. By translating the points in the transition area near the dividing line as shown above, the transition area of ​​the straight line segment can be transformed into the transition area of ​​the curved segment. For the geometric partitioning method implemented by mask, by translating the values ​​of the points on the mask as shown above, the mask of the straight line geometric partitioning can be transformed into the mask of the curved geometric partitioning.

[0152] In some embodiments of the present application, the moving distance is the distance between the corresponding pixel position on the preset curve shape and the straight dividing line; the moving vector is a preset vector corresponding to the pixel position on the preset curve shape.

[0153] It should be noted that, in the embodiment of the present application, a LUT (look up table) may be used to store preset vectors or preset distances corresponding to the curve divisions.

[0154] In an embodiment of the present application, it is assumed that all points move in the same direction (horizontally, vertically, or in the direction of another angle), and the LUT stores the moving distances of several points for a curve shape. One or more curve shapes can be stored in the LUT. Among them, the curve shape is the shape of the curve itself. When a straight line is translated in a certain direction according to the curve direction, the curve may not be the same as the shape of the curve itself. If the divided line segment is to be translated according to a certain curve shape, each point on the line segment finds the corresponding position in the corresponding curve shape in the LUT, and moves in a given direction according to the indicated moving distance. The point can be restricted from moving beyond a certain boundary. If the mask divided by a straight line is to be converted into a mask divided by a curve, after determining the direction of translation, the values ​​of different points on the mask perpendicular to the translation direction are moved according to the moving distances of the corresponding positions on the LUT, and the values ​​of the points on the mask with the same translation direction are moved according to the same moving distance. It can be understood that the new mask divided by the curve is obtained by referring to the mask divided by the straight line.

[0155] In some embodiments of the present application, point values ​​outside the required range are discarded, or point values ​​can be restricted from moving beyond a certain boundary, and missing point values ​​are supplemented with the original values ​​at the edge of the valid range.

[0156] In some embodiments of the present application, the value of a point in the mask divided by the curve is found from the value of the corresponding point in the mask divided by the straight line in the opposite direction and distance it should move. If the point to be found exceeds the valid limit, it is replaced by the value of the point on the limit on the same straight line as the moving direction.

[0157] In some embodiments of the present application, assuming that not all points move in the same direction, the LUT stores the motion vectors of several points for a given curve shape. The LUT can store one or more curve shapes. To translate a segmented line according to a specific curve shape, each point on the segment finds its corresponding position in the corresponding curve shape in the LUT and moves according to the indicated motion vector. Discontinuities caused by the movement may require additional points at the discontinuous locations to maintain curve continuity. To convert a mask divided by straight lines into a mask divided by curved lines, after determining the direction of translation, the values ​​of different points in a certain direction on the mask can be shifted according to the motion vectors at the corresponding positions in the LUT, supplementing the values ​​of the points where the discontinuity occurs. The values ​​of points on the mask in the same direction perpendicular to the aforementioned direction are shifted according to the same motion vector. The values ​​of the points where the discontinuity occurs are supplemented. Point values ​​outside the required range are discarded, or the point values ​​can be constrained to not exceed a certain boundary after movement. The values ​​of missing points are supplemented according to the original values ​​at the edge of the required range.

[0158] In some embodiments of the present application, the moving direction is formed by combining the horizontal direction and the vertical direction; or, the moving direction is related to the division direction of the GPM.

[0159] In some embodiments of the present application, when the moving distance is not an integer pixel, the sub-pixel position is interpolated or rounded; or, the moving distance is positively correlated with the length of the current block; or, the moving distance is positively correlated with the minimum value of the length and width of the current block.

[0160] In some embodiments of the present application, the movement direction may be horizontal or vertical, or any other direction. If the movement direction is not horizontal or vertical, it can be divided into movement in both horizontal and vertical directions.

[0161] For example, a 45° movement of 2 pixels can be split into a horizontal 0° movement pixel distance plus 90° vertical movement Pixel distance.

[0162] In the embodiment of the present application, when the moving distance is not an integer pixel, one possible implementation method is to take an approximate integer pixel position, such as a rounding operation, and another possible implementation method is to interpolate the sub-pixel position.

[0163] For example, The pixel distance can be approximated as 1 pixel, 2 pixels, 1.5 pixels, or 1.4 pixels. The values ​​for 1.5 pixels or 1.4 pixels require interpolation. Other angles can also be split into horizontal and vertical movements based on trigonometric functions or pre-calculated ratios. Horizontal movement can be represented by a positive value for 0° and a negative value for 180°. Vertical movement can be represented by a positive value for 90° and a negative value for 270°.

[0164] In the embodiments of the present application, one possible implementation is to use only the horizontal and vertical movement directions; another possible implementation is to use only the horizontal and vertical movement directions plus the 45°, 135°, 225°, and 315° movement directions. The 45°, 135°, 225°, and 315° movement directions can all be decomposed into equal horizontal and vertical movement directions, and the four angles can be represented by the positive and negative signs of the horizontal and vertical directions; another possible implementation is to use multiple angle directions.

[0165] It should be noted that the direction of movement can be related to the division direction of GPM. In the case of only horizontal and vertical movement, if the straight line angle used for GPM straight line division is horizontal or close to horizontal, the curve division uses the vertical movement direction; if the straight line angle used for GPM straight line division is vertical or close to vertical, the curve division uses the horizontal movement direction. More generally, some thresholds can be set to use the horizontal movement direction within a certain threshold range when the straight line angle or the angle used for division or the division mode is within a certain threshold range, and use the vertical movement direction within another range. The same principle applies to the case of only using the horizontal movement direction plus the vertical movement direction plus 45°, 135°, 225°, 315° movement directions or using multiple movement directions, setting a threshold range for the straight line angle or the angle used for division or the division mode to distinguish the usage range of each movement direction.

[0166] It should be noted that the movement distance may be related to the length and width of the current block. One possible implementation is that the movement distance is proportional to the length of the current block in the same direction as the current movement direction. For example, if the length of the current block in the same direction as the current movement direction is 64, the distance moved by all points in the current block is the value found in the LUT. If the length of the current block in the same direction as the current movement direction is 32, the distance moved by all points in the current block is 1 / 2 of the value found in the LUT. One possible implementation is that the movement distance is proportional to the shorter length of the current block.

[0167] In the embodiment of the present application, different translation amplitudes can be set, i.e., 1 / 4, 1 / 2, 1, 2, etc. If 1 / 4 is selected for the current block, the distance moved by all points in the current block is 1 / 4 times the movement distance found in the LUT. The translation amplitude can be 0, i.e., the distance moved by all points in the current block is 0. In this case, the curve division does not actually work. A negative translation amplitude indicates that the movement direction is reversed from the original direction.

[0168] That is, the translation amplitude is a multiple of the actual moving distance or the moving vector, or a geometric relationship.

[0169] In some embodiments of the present application, not all points to be offset need to be translated in the same direction. Points on a straight line segment can be translated, with each point's translation vector not being the same, and adjacent points' translation vectors being similar, to maintain line continuity and still produce a curved segment.

[0170] For video coding, the so-called straight line is composed of points whose pixel positions are on the same straight line or approximately on the same straight line, and the points are translated in the same direction, and the distance of translation of each point is not the same, and the distance of translation of adjacent points meets certain requirements, and a curve segment can be obtained. The points are translated, and the translation vector of each point is not the same, and the distance of translation of adjacent points meets certain requirements, and a curve segment can be obtained.

[0171] Assuming that not all points are moved in the same direction, the LUT stores the moving vectors of a plurality of points of a curve shape (the LUT can store information related to one or more curve shapes). If the divided line segment is to be translated according to a certain curve shape, each point on the line segment finds the corresponding position in the corresponding curve shape in the LUT, and moves according to the indicated moving vector. Due to the discontinuity caused by the movement, points may need to be supplemented at the discontinuous positions to maintain the continuity of the curve. If the straight line division mask is to be changed into a curve division mask, the values of different points in a certain direction on the mask can be moved according to the moving vectors of the corresponding positions on the LUT after the direction of translation is determined, and the values of the points at the discontinuous positions are supplemented. The values of the points in the same perpendicular direction on the mask as the above direction are moved according to the same moving vectors. The values of the points at the discontinuous positions are supplemented. The values of the points beyond the required range are discarded, or the values of the points can be limited to not exceed a certain boundary after movement, and the values of the missing points are supplemented according to the original values of the edges of the required range.

[0172] In the embodiments of the present application, the LUT is used to determine the distance of translation or the translation vector.

[0173] It can be understood that the curve geometry division can be realized by using the translation method, which can make the division boundary more flexible and more consistent with the edge of the object or the edge of the part with different motion vectors, thereby improving the coding performance. Moreover, the complexity of the curve geometry division realized by the translation method is low, and the translation method is easy to implement.

[0174] The embodiments of the present application provide an inter prediction method, which is applied to a video coding device, i.e., an encoder. The functions realized by the method can be realized by calling program codes by a second processor in the video coding device, and of course the program codes can be stored in a second memory. Therefore, the video coding device at least includes the second processor and the second memory.

[0175] Figure 7 For the implementation flowchart of the inter prediction method of the embodiments of the present application, as shown in Figure 7 the method comprises:

[0176] S201, determining a prediction mode parameter of a current block.

[0177] S202: When the prediction mode parameter indicates that the inter-frame geometric partitioning prediction mode GPM is used to determine the inter-frame prediction of the current block, determine the GPM mode parameters and curve partitioning information of the current block.

[0178] S203. Determine a first reference block and a second reference block based on the GPM mode parameters.

[0179] S204 : Determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve division information.

[0180] S205 . Determine a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

[0181] In an embodiment of the present application, a video image can be divided into multiple image blocks, each image block currently to be encoded can be called a coding block (Coding Block, CB), wherein each coding block may include a first image component, a second image component, and a third image component; and the current block is a coding block in the video image that is currently to be predicted for the first image component, the second image component, or the third image component.

[0182] Here, assuming that the current block performs a first image component prediction, and the first image component is a luminance component, that is, the image component to be predicted is a luminance component, then the current block can also be called a luminance block; or, assuming that the current block performs a second image component prediction, and the second image component is a chrominance component, that is, the image component to be predicted is a chrominance component, then the current block can also be called a chrominance block.

[0183] In some embodiments of the present application, the encoder determines the prediction mode parameters of the current block as follows: the encoder determines the image component to be predicted of the current block; based on the parameters of the current block, uses multiple prediction modes to perform predictive encoding on the image component to be predicted respectively, and calculates the rate-distortion cost results corresponding to each prediction mode under the multiple prediction modes; selects the minimum rate-distortion cost result from the multiple rate-distortion cost results calculated, and determines the prediction mode corresponding to the minimum rate-distortion cost result as the prediction mode parameter of the current block.

[0184] It should be noted that the prediction mode parameters indicate the coding mode of the current block and parameters related to the mode. Usually, the prediction mode parameters of the current block can be determined by using Rate Distortion Optimization (RDO).

[0185] In some embodiments of the present application, the encoder determines the prediction mode parameters of the current block as follows: the encoder determines the image component to be predicted of the current block; based on the parameters of the current block, uses multiple prediction modes to perform predictive encoding on the image component to be predicted respectively, and calculates the rate-distortion cost results corresponding to each prediction mode under the multiple prediction modes; selects the minimum rate-distortion cost result from the multiple rate-distortion cost results calculated, and determines the prediction mode corresponding to the minimum rate-distortion cost result as the prediction mode parameter of the current block.

[0186] That is, on the encoder side, multiple prediction modes can be used for the current block to encode the predicted image components separately. Here, the multiple prediction modes generally include inter-frame prediction mode, traditional intra-frame prediction mode and non-traditional intra-frame prediction mode, and the traditional intra-frame prediction mode can include direct current (DC) mode, planar (PLANAR) mode and angle mode, etc., and the non-traditional intra-frame prediction mode can include MIP mode, cross-component linear model prediction (CCLM) mode, intra block copy (IBC) mode and PLT (Palette) mode, etc. The inter-frame prediction mode can include: inter-frame geometric partition prediction mode (GPM), triangle prediction mode (TPM), etc. In the embodiment of the present application, a curve partitioning mode is also included, wherein, in the embodiment of the present application, the LUT stores all curve partitioning information of the curve partitioning.

[0187] In this way, after encoding the current block using multiple prediction modes respectively, the rate-distortion cost results corresponding to each prediction mode can be obtained; then the minimum rate-distortion cost result is selected from the multiple rate-distortion cost results obtained, and the prediction mode corresponding to the minimum rate-distortion cost result is determined as the prediction mode parameter of the current block; in this way, the current block can finally be encoded using the determined prediction mode, and under this prediction mode, the prediction residual can be made small, which can improve the coding efficiency.

[0188] In the embodiment of the present application, since there can be many prediction modes, and a prediction mode can have many specific division modes, it is necessary to first traverse all prediction modes, and then traverse the division modes for each prediction mode, etc. The multiple prediction modes in the embodiment of the present application include the GPM mode, and the present application is implemented based on the scenario of the GPM prediction mode.

[0189] In an embodiment of the present application, when multiple prediction modes include a GPM mode, the process of the encoder calculating the rate-distortion cost result corresponding to the GPM prediction mode can be: the encoder obtains a set of motion-related information corresponding to the current block; uses each GPM partition mode in the GPM partition mode to traverse each motion-related information in the motion-related information set, and then uses the curve partition mode to traverse each GPM mode to obtain the rate-distortion cost result corresponding to the GPM prediction mode.

[0190] It should be noted that each motion-related information may be two MVs, which is not limited in the embodiment of the present application. The motion-related information is stored in a merge list.

[0191] In an embodiment of the present application, if the encoder uses a mask method when performing GPM division, the encoder will determine whether the current block uses GPM and curve division. One possible method is to traverse all possible mv combinations (motion-related information sets) for the current block, traverse all possible GPM division modes for each mv combination, and traverse all curve division modes for each GPM division mode, including all translation amplitudes and curve shapes. For each possible situation, the actual coding cost (rate-distortion cost result) is obtained, and the GPM division mode and curve division mode corresponding to the best coding cost (i.e., the minimum rate-distortion cost result) are taken as the GPM mode and curve division mode of the current block.

[0192] In an embodiment of the present application, for each of the above possible situations, the cost is cost=SAD+bitcost or cost=SATD+bitcost as the encoding cost, where SAD is the Sum of Absolute Difference, SATD is the Sum of Absolute Transformed Difference, and bitcost is the cost of the encoding flag bit, etc.

[0193] In some embodiments of the present application, when multiple prediction modes include a GPM mode, the process of the encoder calculating the rate-distortion cost result corresponding to the GPM prediction mode can be: the encoder obtains a motion-related information set corresponding to the current block; adopts the GPM partitioning mode, traverses each motion-related information in the motion-related information set, and obtains a first rate-distortion cost result corresponding to the GPM partitioning mode; for at least one target GPM partitioning mode with the lowest first rate-distortion cost result, traverses the curve partitioning mode to obtain the rate-distortion cost result corresponding to the GPM prediction mode.

[0194] In an embodiment of the present application, another possible method is that the encoder traverses all possible mv combinations for the current block, traverses all possible GPM division modes for each mv combination, and uses cost = SAD + bitcost or cost = SATD + bitcost as the encoding cost for each possible case. The GPM division modes with the best encoding costs are selected, and all curve division modes are traversed for each GPM division mode, including all translation amplitudes and curve shapes. For each possible case, the cost = SAD + bitcost or cost = SATD + bitcost is used as the encoding cost. The GPM division mode and curve division mode with the best encoding cost are selected as the GPM mode and curve division mode of the current block.

[0195] It should be noted that the implementation principle of each GPM division mode traversing all curve division modes is consistent with the principle of using curve division information for offset on the decoding side, and will not be repeated here.

[0196] In an embodiment of the present application, the encoder determines the optimal encoding cost, i.e., the GPM mode parameters and curve partitioning information corresponding to the current block when the rate-distortion cost result is obtained. Based on the GPM mode parameters, the encoder can determine a first reference block and a second reference block corresponding to the motion-related information. Based on the GPM mode parameters and the curve partitioning information, the encoder can determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block. Finally, the encoder determines a prediction value for the current block based on the first prediction parameter and the second prediction parameter.

[0197] In an embodiment of the present application, the first prediction parameter includes: a first weight value or first boundary position information; the second prediction parameter includes: a second weight value or second boundary position information.

[0198] It should be noted that the first weight value and the second weight value are weight values ​​corresponding to pixel points when GPM division is performed using the mask method, and the first boundary position information and the second boundary position information are boundary information of the first reference block and the second reference block.

[0199] In the embodiment of the present application, the encoder determines the first prediction parameter corresponding to the first reference block and the second prediction parameter corresponding to the second reference block based on the GPM mode parameter and the curve partitioning information. The determination can be divided into two ways based on the difference in the prediction parameters, as follows:

[0200] Method 1: The first prediction parameter includes: a first weight value; the second prediction parameter includes: a second weight value; the encoder determines a preset weight distribution corresponding to the current GPM mode based on the GPM mode parameter; determines the size information of the current block; updates the preset weight distribution based on the size information, the GPM mode parameter and the curve division information to obtain a curve weight distribution; obtains the first weight value and the second weight value based on the curve weight distribution.

[0201] It should be noted that the curve division information includes: translation amplitude and curve shape index; the encoder determines the offset direction according to the GPM mode parameters; determines the offset amount according to the size information, translation amplitude and curve shape index; according to the offset direction, the preset weight of each pixel position in the preset weight distribution is offset by the offset amount to obtain the updated curve weight of each pixel position, thereby obtaining the curve weight distribution.

[0202] Method 2: The first prediction parameter includes: first boundary position information; the second prediction parameter includes: second boundary position information; the encoder obtains the size information of the current block; based on the GPM mode parameter and the size information, the division boundary position information is determined; according to the GPM mode parameter, the size information and the curve division information, the division boundary position information is offset to determine the curve division boundary position information; based on the curve division boundary position information, the first boundary position information and the second boundary position information are determined.

[0203] It should be noted that the curve division information includes: translation amplitude and curve shape index; the encoder determines the offset direction according to the GPM mode parameters; determines the offset amount according to the size information, translation amplitude and curve shape index; according to the offset direction, the position of each pixel point of the division boundary position information is offset by the offset amount to obtain the curve division boundary position information.

[0204] In Method 1, the preset weight distribution refers to the original mask corresponding to the GPM division mode using straight-line division. The curved weight distribution is the weight distribution information corresponding to the current curved division, which is updated by performing a translation calculation on the preset weight distribution. The encoder primarily uses size information, GPM mode parameters, and curved division information to update the preset weight distribution to obtain the curved weight distribution. From the curved weight distribution, the first weight value corresponding to the first reference block and the second weight value corresponding to the second reference block are obtained for each pixel point of the current block.

[0205] In some embodiments of the present application, the encoder determines the offset direction based on the angle information in the GPM mode parameters, and determines the offset amount required based on the size information, the translation amplitude, and the curve shape index.

[0206] In method 2, the encoder adopts a curve division method for the current block, and can use the pixel value of the first reference block as the first prediction value for part of the first reference block, and use the pixel value of the second reference block as the second prediction value for part of the second reference block. However, at the boundary position, the pixel points on the respective boundary information of the first reference block and the second reference block are offset to obtain the first boundary position information and the second boundary position information, and then the pixel points on the boundary are predicted based on the first boundary position information and the second boundary position information.

[0207] It should be noted that the encoder uses the GPM mode parameters to divide the current block. Based on the size information of the current block, the division boundary position information of the current block can be obtained; at this time, the encoder determines the offset direction according to the GPM mode parameters; determines the offset amount according to the size information, translation amplitude and curve shape index; according to the offset direction, the position of each pixel point of the division boundary position information is offset by the offset amount to obtain the curve division boundary position information.

[0208] In response to the above implementation, in an embodiment of the present application, the encoder determines the offset direction based on the GPM mode parameters as follows: when the angle used by the GPM represented by the GPM mode parameters meets the horizontal judgment condition, the offset direction is determined to be the horizontal direction; when the angle used by the GPM represented by the GPM mode parameters meets the vertical judgment condition, the offset direction is determined to be the vertical direction.

[0209] In the following, for different offset modes, the encoder determines the description of the offset based on the size information, translation amplitude and curve shape index.

[0210] In the embodiment of the present application, the size information includes the length and width of the current block.

[0211] When the offset direction is horizontal, the encoder determines the first pixel position in the preset block in the non-offset direction at each pixel position based on each pixel position, the length, and the number of preset offset points, where the non-offset direction is perpendicular to the offset direction; determines first curve information corresponding to the first pixel position based on the curve shape index; determines a first movement ratio based on the width and the number of preset offset points; and determines an offset amount based on the first curve information, the translation amplitude, and the first movement ratio, where the offset amount does not exceed the width of the current block.

[0212] In some embodiments of the present application, the encoder determines the first pixel position in the preset block at each pixel position in the non-offset direction based on each pixel position, length, and the number of preset offset points as follows: determining a first scaling ratio based on each pixel position, length, and the number of preset offset points; determining a first sub-offset based on the length and the number of preset offset points; and determining the first pixel position in the preset block at each pixel position in the non-offset direction based on the first scaling ratio and the first sub-offset.

[0213] In some embodiments of the present application, when the offset direction is a vertical direction, the encoder determines a second pixel position in a preset block in a non-offset direction at each pixel position based on each pixel position, width, and the number of preset offset points, wherein the non-offset direction is perpendicular to the offset direction; determines second curve information corresponding to the second pixel position based on a curve shape index; determines a second movement ratio based on the length and the number of preset offset points; and determines an offset amount based on the second curve information, the translation amplitude, and the second movement ratio, wherein the offset amount does not exceed the length of the current block.

[0214] In the embodiment of the present application, the encoder multiplies the second curve information with the translation amplitude and the second movement ratio to obtain an offset.

[0215] In some embodiments of the present application, the encoder determines the second pixel position located in the preset block in the non-offset direction at each pixel position based on each pixel position, the width, and the number of preset offset points as follows: the encoder determines a second scaling ratio based on each pixel position, the width, and the number of preset offset points; determines a second sub-offset based on the width and the number of preset offset points; and determines the second pixel position located in the preset block in the non-offset direction at each pixel position based on the second scaling ratio and the second sub-offset.

[0216] Furthermore, after the encoder obtains the first weight value and the second weight value using method 1, the encoder can determine the prediction value of the current block based on the first weight value, the first pixel value of the first reference block, the second weight value and the second pixel value of the second reference block.

[0217] In an embodiment of the present application, the encoder weights and averages the first weight value, the first pixel value of the first reference block, the second weight value, and the second pixel value of the second reference block to obtain a prediction value of the current block.

[0218] After the encoder obtains the first boundary position information and the second boundary position information using method 2, the encoder divides the current block according to the first boundary position information and the second boundary position information to obtain a first block and a second block; predicts the first block based on the first pixel value of the first reference block, and predicts the second block based on the second pixel value of the second reference block, thereby determining a predicted value of the current block.

[0219] The inter-frame prediction provided in the embodiment of the present application adopts a curve division method.

[0220] In an embodiment of the present application, the curve division information is a division method in curve division; curve division is to translate the pixel points on the direct division line in the preset straight line geometric division to obtain a continuous curve division line to realize geometric division.

[0221] In some embodiments of the present application, each pixel point of a straight line dividing line is continuously translated according to different moving distances and the same moving direction to obtain a curved line dividing line; or, each pixel point of a straight line dividing line is continuously translated according to different moving vectors to obtain a curved line dividing line.

[0222] In some embodiments of the present application, the moving distance is the distance between the corresponding pixel position on the preset curve shape and the straight dividing line; the moving vector is a preset vector corresponding to the pixel position on the preset curve shape.

[0223] In some embodiments of the present application, the moving direction is formed by combining the horizontal direction and the vertical direction; or, the moving direction is related to the division direction of the GPM.

[0224] In some embodiments of the present application, when the moving distance is not an integer pixel, the sub-pixel position is interpolated or rounded; or, the moving distance is positively correlated with the length of the current block; or, the moving distance is positively correlated with the minimum value of the length and width of the current block.

[0225] It should be noted that the principle of the prediction process of the current block of the encoder is consistent with that of the decoder side, and the same explanation and description will not be repeated here.

[0226] In some embodiments of the present application, if the encoder uses the CU or PU partitioning method when performing GPM partitioning, the encoder needs to determine whether the current block uses the GPM partitioning mode (or GPM prediction mode) and the curve partitioning mode.

[0227] In the embodiments of the present application, the encoder can traverse all GPM partition modes and curve partition modes, and encode the divided CUs or PUs for each possible case. The process can include an iterative encoding process of the CUs or PUs in the next layer to obtain the encoding cost. The encoder attempts other methods that do not use GPM and curve partition to encode the current block, and determines the prediction mode selected by the current block according to the mode with the minimum encoding cost.

[0228] It can be understood that the encoder predicts the current block by using the GPM mode and the curve partition mode, so that the divided boundary is more consistent with the edge of the object or the edge of the part with different motion vectors, thereby improving the encoding accuracy.

[0229] In some embodiments of the present application, after S202, as shown in the following S206-S207, the inter prediction method provided by the present application further includes the following steps. Figure 8

[0230] S206, when the prediction mode corresponding to the minimum rate-distortion cost result is the curve partition mode of the GPM prediction mode, generating a curve partition usage flag according to the curve partition information, and encoding the curve partition information.

[0231] S207, writing the curve partition usage flag and the encoded curve partition information into a bitstream.

[0232] In the embodiments of the present application, after the encoder determines the minimum rate-distortion cost result, if the prediction mode corresponding to the minimum rate-distortion cost result is the curve partition mode of the GPM prediction mode, the encoder can generate a curve partition usage flag according to the curve partition information, and encode the curve partition information. Finally, the encoder writes the curve partition usage flag and the encoded curve partition information into a bitstream for use by the decoder when decoding. However, in the present application, only the encoded curve partition information can be written into the bitstream for use by the decoder when decoding, and the embodiments of the present application are not limited in this regard.

[0233] In some embodiments of the present application, the curve partition information includes a translation amplitude and a curve shape index. The encoder can generate a curve partition usage flag according to the translation amplitude, encode the translation amplitude and the curve shape index when the curve partition usage flag indicates that the curve partition is used, and obtain the encoded curve partition information. When the curve partition usage flag indicates that the curve partition is not used, the encoding of the translation amplitude and the curve shape index is stopped.

[0234] ​In some embodiments of the present application, the encoder may further encode the translation amplitude to obtain the encoded translation amplitude; when the translation amplitude is represented by using curve division, the curve shape index is encoded to obtain the encoded curve shape index, wherein the encoded translation amplitude and the encoded curve shape index are the encoded curve division information; when the translation amplitude is represented by using curve division and the representation does not use curve division, the encoding of the curve shape index is stopped.

[0235] It should be noted that in the embodiment of the present application, the encoder can use the GPM usage flag or other flags to derive whether GPM is used. If the current block uses the GPM mode, the information of the selected GPM partitioning mode is encoded. The method of encoding the GPM mode is not limited in this application.

[0236] In this embodiment of the present application, the encoder needs to determine the GPM usage flag or derive whether GPM is used through other flags. If the current block uses the selected GPM mode, the encoder encodes the selected GPM partitioning mode information, the MV information used by the two partitions of the current block, or other GPM-related information. The method of encoding the GPM mode is not limited in this application.

[0237] In the embodiment of the present application, for the curve division mode, a curve division use flag indicates whether the curve division method is used. The curve division use flag can be determined based on the translation amplitude. When the translation amplitude is present, the curve division use flag is generated as a used flag. When the translation amplitude is absent or the translation amplitude is less than a certain value, the curve division use flag is generated as a non-used flag, and no curve division information is encoded. The translation amplitude is determined based on the movement distance and the movement vector.

[0238] For example, if the translation amplitude is not 0, the curve segmentation usage flag is set to 1, and the translation amplitude and curve shape index are encoded. The encoding order of the translation amplitude and curve shape index is not restricted. If the translation amplitude is not 0, the curve segmentation usage flag is set to 0, and the encoding of the curve segmentation information stops.

[0239] In some embodiments of the present application, the encoder may first encode the translation amplitude. If the translation amplitude is not 0 (such as 1), it indicates that curve division is used and the encoding of the curve shape index continues. If the translation amplitude is 0, the encoding of the curve division information stops.

[0240] It should be noted that the order of encoding the curve division information, the GPM division mode, the MV information used in the two divisions of the current block, or other GPM-related information may be combined in various ways, and this application does not impose any restrictions. This application does not impose any restrictions on the encoding of the GPM division mode, the MV information used in the two divisions of the current block, or other GPM-related information.

[0241] In other words, how to determine whether to use curve partitioning for the current block under the premise of using GPM? One possible implementation is to use a curve partitioning use flag. In this method, the range of the translation amplitude does not include 0. Another possible implementation is to use the translation amplitude. If the translation amplitude is 0, the curve partitioning is not actually used.

[0242] It should be noted that, in the embodiment of the present application, curve division can use a variety of curve shapes, and each curve shape can be represented as an array in the LUT (that is, all curve division information is stored). If it is determined to use curve division, it is necessary to use the curve shape index to determine which curve shape to use. The curve shapes that can be used for curve division may be different depending on the GPM division mode, and the curve shape index is an index in the available shapes. If curve division cannot be used in a certain GPM division mode, then the curve division usage flag, translation amplitude, and curve shape index, etc. do not need to be used in this GPM mode. If only one curve shape can be used in a certain GPM division mode, then the curve shape index may not be needed in this GPM mode.

[0243] It is understandable that when the encoder adopts the curve partitioning mode to encode the current block, parameters related to the curve partitioning can be written into the bitstream for use during decoding, thereby improving decoding efficiency.

[0244] The following takes VVC draft 8 as an example to describe the inter-frame prediction part provided by the embodiments of the present application. The underlined part below is the embodiment of the main improvements.

[0245] Merge data syntax

[0246]

[0247]

[0248] Weighted sample prediction process for geometric partitioning mode

[0249] Inputs to this process are:

[0250] -two variables nCbW and nCbH specifying the width and the height of the current coding block,

[0251] -two(nCbW)x(nCbH)arrays predSamplesLA and predSamplesLB,

[0252] -a variable angleIdx specifying the angle index of the geometricpartition,

[0253] -a variable distanceIdx specifying the distance index of thegeometric partition,

[0254] -a variable cIdx specifying colour component index,

[0255] - avariablecurveAmplitudespecifyingcurveamplitude,

[0256] - avariablecurveIdx specifying curve index .

[0257] Output of this process is the(nCbW)x(nCbH)array pbSamples ofprediction sample values.

[0258] The variables nW,nH,shift1,offset1,hwRatio,displacementX,displacementY,partFlip and shiftHor arederived as follows:

[0259] nW = ( cIdx == 0 ) ? nCbW : nCbW * SubWidthC (1030)

[0260] nH = ( cIdx == 0 ) ? nCbH : nCbH * SubHeightC (1031)

[0261] shift1 = Max(5,17-BitDepth) (1032)

[0262] offset1 = 1<<(shift1-1) (1033)

[0263] hwRatio = nH / nW (1034)

[0264] displacementX = angleIdx (1035)

[0265] displacementY = (angleIdx+8)%32 (1036)

[0266] partFlip = ( angleIdx >= 13 && angleIdx <= 27 ) ? 0 : 1 (1037)

[0267] shiftHor = ( angleIdx%16 == 8 ||(angleIdx%16!=0 && hwRatio > 0 ) )? 0 : 1 (1038)

[0268] The variables offsetX and offsetY are derived as follows:

[0269] -If shiftHor is equal to 0,the following applies:

[0270] offsetX = (-nW)>>1 (1039)

[0271] offsetY=((-nH)>>1)+

[0272] (angleIdx<16?(distanceIdx*nH)>>3:-((distanceIdx*nH)>>3)) (1040)

[0273] -Otherwise(shiftHor is equal to 1),the following applies:

[0274] offsetX=((-nW)>>1)+

[0275] (angleIdx<16?(distanceIdx*nW)>>3:-((distanceIdx*nW)>>3)) (1041)

[0276] offsetY = (-nH)>>1 (1042)

[0277] The prediction samples pbSamples[x][y]with x=0..nCbW-1and y=0..nCbH-1are derived as follows:

[0278] -The variables xL and yL are derived as follows:

[0279] xL=(cIdx==0)?x:x*SubWidthC(1043)

[0280] yL=(cIdx==0)?y:y*SubHeightC(1044)

[0281] - The variables offsetCX and offsetCY are derived as follows:

[0282] If((angleIdx>=4&&angleIdx<12)||(angleIdx>=20&&angleIdx<28)), the following applies:

[0283] offsetCX=0

[0284] offsetCY=curveFactor[curveIdx][xL*N / nW+N / 2nW]*curveAmplitude*nH / N

[0285] - Otherwise(shiftHorisequalto1),thefollowingapplies:

[0286] offsetCX=curveFactor[curveIdx][yL*N / nH+N / 2nH]*curveAmplitude*nW / N

[0287] offsetCY=0

[0288] -The variable wValue specifying the weight of the prediction sampleis derived based on the arraydisLut specified in Table37 as follows:

[0289] weightIdx=(((xL+offsetX+ offsetCX )<<1)+1)*disLut[displacementX]+

[0290] (((yL+offsetY+ offsetCY )<<1)+1))*disLut[displacementY](1045)

[0291] weightIdxL=partFlip?32+weightIdx:32-weightIdx(1046)

[0292] wValue=Clip3(0,8,(weightIdxL+4)>>3)(1047)

[0293] -The prediction sample values ​​are derived as follows:

[0294] pbSamples[x][y]=Clip3(0,(1< <BitDepth)-1,(predSamplesLA[x][y]*wValue+(1048)

[0295] predSamplesLB[x][y]*(8-wValue)+offset1)>>shift1)

[0296] Table37-Specification of the geometric partitioning distance arraydisLut.

[0297] idx 0 2 3 4 5 6 8 10 11 12 13 14 disLut[idx] 8 8 8 4 4 2 0 -2 -4 -4 -8 -8 idx 16 18 19 20 21 22 24 26 27 28 29 30 disLut[idx] -8 -8 -8 -4 -4 -2 0 2 4 4 8 8

[0298] A possible implementation is N=64.

[0299] like Figure 9 As shown, an embodiment of the present application provides a decoder 1, comprising:

[0300] The decoding unit 10 is used to parse the code stream and determine the prediction decoding parameters of the current block;

[0301] A first determining unit 11 is configured to, when the prediction decoding parameter indicates that an inter-frame geometric partitioning prediction mode GPM is used to determine inter-frame prediction of the current block, determine GPM mode parameters and curve partitioning information of the current block; determine a first reference block and a second reference block based on the GPM mode parameters; and determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve partitioning information;

[0302] The first prediction unit 12 is configured to determine a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

[0303] In some embodiments of the present application, the first prediction parameter includes: a first weight value or first boundary position information; the second prediction parameter includes: a second weight value or second boundary position information.

[0304] In some embodiments of the present application, the first prediction parameter includes: a first weight value; the second prediction parameter includes: a second weight value;

[0305] The first determination unit 11 is further used to determine a preset weight distribution corresponding to the current GPM mode based on the GPM mode parameters; determine the size information of the current block; update the preset weight distribution based on the size information, the GPM mode parameters and the curve division information to obtain a curve weight distribution; and obtain the first weight value and the second weight value based on the curve weight distribution.

[0306] In some embodiments of the present application, the curve division information includes: a translation amplitude and a curve shape index; the first determination unit 11 is also used to determine the offset direction according to the GPM mode parameters; determine the offset amount according to the size information, the translation amplitude and the curve shape index; according to the offset direction, the preset weight of each pixel position in the preset weight distribution is offset by the offset amount to obtain the updated curve weight of each pixel position, thereby obtaining the curve weight distribution.

[0307] In some embodiments of the present application, the first prediction unit 12 is further configured to determine a prediction value of the current block by combining the first weight value, the first pixel value of the first reference block, the second weight value and the second pixel value of the second reference block.

[0308] In some embodiments of the present application, the first prediction parameter includes: first boundary position information; the second prediction parameter includes: second boundary position information;

[0309] The first determination unit 11 is also used to obtain the size information of the current block; determine the division boundary position information based on the GPM mode parameters and the size information; offset the division boundary position information according to the GPM mode parameters, the size information and the curve division information to determine the curve division boundary position information; determine the first boundary position information and the second boundary position information based on the curve division boundary position information.

[0310] In some embodiments of the present application, the first prediction unit 12 is further used to divide the current block according to the first boundary position information and the second boundary position information to obtain a first block and a second block; predict the first block based on the first pixel value of the first reference block, and predict the second block based on the second pixel value of the second reference block, so as to determine the predicted value of the current block.

[0311] In some embodiments of the present application, the curve division information comprises a translation amplitude and a curve shape index; the first determining unit 11 is further configured to determine a shift direction according to the GPM mode parameter; determine a shift amount according to the size information, the translation amplitude and the curve shape index; and obtain the curve division boundary position information by shifting the position of each pixel point of the division boundary position information by the shift amount in the shift direction.

[0312] In some embodiments of the present application, the first determining unit 11 is further configured to determine that the shift direction is a horizontal direction when the GPM mode parameter represents an angle used by GPM satisfying a horizontal judgment condition; and determine that the shift direction is a vertical direction when the GPM mode parameter represents an angle used by GPM satisfying a vertical judgment condition.

[0313] In some embodiments of the present application, the size information comprises a length and a width of a current block.

[0314] The first determining unit 11 is further configured to, when the shift direction is a horizontal direction, determine a first pixel position in a preset block in a non-shift direction at the each pixel position according to the each pixel position, the length and a preset number of shift points, wherein the non-shift direction is perpendicular to the shift direction; determine first curve information corresponding to the first pixel position according to the curve shape index; determine a first moving ratio based on the width and the preset number of shift points; and determine the shift amount according to the first curve information, the translation amplitude and the first moving ratio, wherein the shift amount does not exceed the width of the current block.

[0315] In some embodiments of the present application, the first determining unit 11 is further configured to determine a first scaling ratio according to the each pixel position, the length and the preset number of shift points; determine a first sub-shift amount according to the length and the preset number of shift points; and determine the first pixel position in a preset block in a non-shift direction at the each pixel position based on the first scaling ratio and the first sub-shift amount.

[0316] In some embodiments of the present application, the size information comprises a length and a width of a current block.

[0317] The first determination unit 11 is further configured to, when the offset direction is a vertical direction, determine, based on each pixel position, the width, and the number of preset offset points, a second pixel position in the preset block in a non-offset direction at each pixel position, wherein the non-offset direction is perpendicular to the offset direction; determine, based on the curve shape index, second curve information corresponding to the second pixel position; determine a second movement ratio based on the length and the number of preset offset points; and determine an offset amount based on the second curve information, the translation amplitude, and the second movement ratio, wherein the offset amount does not exceed the length of the current block.

[0318] In some embodiments of the present application, the first determination unit 11 is further configured to determine a second scaling ratio based on each pixel position, the width, and the number of preset offset points; determine a second sub-offset based on the width and the number of preset offset points; and determine, based on the second scaling ratio and the second sub-offset, that each pixel position is located at the second pixel position in the preset block in a non-offset direction.

[0319] In some embodiments of the present application, the curve division information is a division method in curve division;

[0320] The curve division is to translate the pixel points on the direct division line in the preset straight line geometric division to obtain a continuous curve division line to achieve geometric division.

[0321] In some embodiments of the present application, each pixel point of the straight dividing line is continuously translated according to different moving distances and the same moving direction to obtain the curved dividing line; or,

[0322] Each pixel point of the straight line dividing line is continuously translated according to different movement vectors to obtain the curved line dividing line.

[0323] In some embodiments of the present application, the movement distance is the distance between the corresponding pixel position on the preset curve shape and the straight dividing line;

[0324] The movement vector is a preset vector corresponding to a pixel position on a preset curve shape.

[0325] In some embodiments of the present application, the moving direction is formed by a combination of a horizontal direction and a vertical direction; or,

[0326] The moving direction is related to the division direction of the GPM.

[0327] In some embodiments of the present application, when the moving distance is not an integer pixel, the sub-pixel position is interpolated or rounded; or,

[0328] The moving distance is positively correlated with the length of the current block; or,

[0329] The moving distance is positively correlated with the smallest value between the length and width of the current block.

[0330] It is understandable that the decoder predicts the current block by using curve partitioning in the GPM mode, so that the partitioned boundary is more closely aligned with the edge of the object or the edge of two parts with different motion vectors, thereby improving the accuracy of encoding and decoding.

[0331] In practical applications, such as Figure 10 As shown, an embodiment of the present application also provides a decoder, including: a first processor 13 and a first memory 14 storing executable instructions of the first processor 13, the first memory 14 relies on the first processor 13 to perform operations through a communication bus 15, and when the executable instructions are executed by the first processor 13, the inter-frame prediction method described on the decoder side is executed.

[0332] In which, the first processor can be implemented by software, hardware, firmware or a combination thereof, and can use circuits, single or multiple application specific integrated circuits (ASICs), single or multiple general-purpose integrated circuits, single or multiple microprocessors, single or multiple programmable logic devices, or a combination of the aforementioned circuits or devices, or other suitable circuits or devices, so that the first processor can execute the corresponding steps of the inter-frame prediction method on the decoder side in the aforementioned embodiment.

[0333] An embodiment of the present application provides a computer-readable storage medium storing executable instructions. When the executable instructions are executed by one or more first processors, the first processors execute an inter-frame prediction method on the decoder side.

[0334] The various components in the embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of software functional modules.

[0335] If the integrated unit is implemented in the form of a software functional module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in this embodiment. The aforementioned storage media include: ferromagnetic random access memory (FRAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic surface storage, optical disk, or compact disc read-only memory (CD-ROM), etc. Various media that can store program codes are not limited in the embodiments of the present disclosure.

[0336] like Figure 11 As shown, the embodiment of the present application further provides an encoder 2, comprising:

[0337] A second determining unit 20 is configured to determine prediction mode parameters of a current block; when the prediction mode parameters indicate that an inter-frame geometric partitioning prediction mode (GPM) is used to determine inter-frame prediction of the current block, determine the GPM mode parameters and curve partitioning information of the current block; determine a first reference block and a second reference block based on the GPM mode parameters; and determine a first prediction parameter corresponding to the first reference block and a second prediction parameter corresponding to the second reference block based on the GPM mode parameters and the curve partitioning information;

[0338] The second prediction unit 21 is configured to determine a prediction value of the current block based on the first prediction parameter and the second prediction parameter.

[0339] In some embodiments of the present application, the encoder further includes: an encoding unit 22;

[0340] The encoding unit 22 is further configured to, after determining the GPM mode parameters, curve division information, and motion-related information of the current block, generate the curve division usage flag according to the curve division information when the prediction mode corresponding to the minimum rate-distortion cost result is the curve division mode of the GPM prediction mode, and encode the curve division information; and input the curve division usage flag and the encoded curve division information into the bitstream.

[0341] In some embodiments of the present application, the curve division information includes: a translation amplitude and a curve shape index; the encoding unit 22 is further used to generate the curve division usage flag based on the translation amplitude; when the curve division usage flag representation is used, the translation amplitude and the curve shape index are encoded to obtain the encoded curve division information; when the curve division usage flag representation is not used, the encoding of the translation amplitude and the curve shape index is stopped.

[0342] In some embodiments of the present application, the curve segmentation information includes: a translation amplitude and a curve shape index; the encoding unit 22 is further configured to encode the translation amplitude to obtain an encoded translation amplitude;

[0343] When the translation amplitude is represented by using curve division, the curve shape index is encoded to obtain the encoded curve shape index, wherein the encoded translation amplitude and the encoded curve shape index are the encoded curve division information; when the translation amplitude is represented by using curve division and the curve division is not used, the encoding of the curve shape index is stopped.

[0344] In some embodiments of the present application, the first prediction parameter includes: a first weight value or first boundary position information;

[0345] The second prediction parameter includes: a second weight value or second boundary position information.

[0346] In some embodiments of the present application, the first prediction parameter includes: a first weight value; the second prediction parameter includes: a second weight value;

[0347] The second determination unit 20 is further used to determine a preset weight distribution corresponding to the current GPM mode based on the GPM mode parameters; determine the size information of the current block; update the preset weight distribution based on the size information, the GPM mode parameters and the curve division information to obtain a curve weight distribution; and obtain the first weight value and the second weight value based on the curve weight distribution.

[0348] In some embodiments of the present application, the curve division information includes: a translation amplitude and a curve shape index; the second determination unit 20 is also used to determine the offset direction according to the GPM mode parameters; determine the offset amount according to the size information, the translation amplitude and the curve shape index; according to the offset direction, the preset weight of each pixel position in the preset weight distribution is offset by the offset amount to obtain the updated curve weight of each pixel position, thereby obtaining the curve weight distribution.

[0349] In some embodiments of the present application, the second prediction unit 21 is further used to determine the prediction value of the current block by combining the first weight value, the first pixel value of the first reference block, the second weight value and the second pixel value of the second reference block.

[0350] In some embodiments of the present application, the first prediction parameter includes: first boundary position information; the second prediction parameter includes: second boundary position information; the second determination unit 20 is also used to obtain the size information of the current block; based on the GPM mode parameter and the size information, determine the division boundary position information; according to the GPM mode parameter, the size information and the curve division information, offset the division boundary position information to determine the curve division boundary position information; based on the curve division boundary position information, determine the first boundary position information and the second boundary position information.

[0351] In some embodiments of the present application, the second prediction unit 21 is further configured to divide the current block into a first block and a second block according to the first boundary position information and the second boundary position information;

[0352] The first block is predicted based on a first pixel value of a first reference block, and the second block is predicted based on a second pixel value of a second reference block, thereby determining a prediction value of the current block.

[0353] In some embodiments of the present application, the curve division information includes: a translation amplitude and a curve shape index; the second determination unit 20 is also used to determine the offset direction according to the GPM mode parameters; determine the offset amount according to the size information, the translation amplitude and the curve shape index; according to the offset direction, the position of each pixel point of the division boundary position information is offset by the offset amount to obtain the curve division boundary position information.

[0354] In some embodiments of the present application, the second determination unit 20 is further used to determine that the offset direction is a horizontal direction when the angle used by the GPM mode parameter to represent the GPM meets a horizontal judgment condition; and to determine that the offset direction is a vertical direction when the angle used by the GPM mode parameter to represent the GPM meets a vertical judgment condition.

[0355] In some embodiments of the present application, the size information includes the length and width of the current block; the second determination unit 20 is further used to determine, when the offset direction is a horizontal direction, the first pixel position in the preset block at each pixel position in a non-offset direction according to each pixel position, the length and the number of preset offset points, wherein the non-offset direction is perpendicular to the offset direction; determine, according to the curve shape index, first curve information corresponding to the first pixel position; determine a first movement ratio based on the width and the number of preset offset points; determine the offset amount according to the first curve information, the translation amplitude and the first movement ratio, wherein the offset amount does not exceed the width of the current block.

[0356] In some embodiments of the present application, the second determination unit 20 is further configured to determine a first scaling ratio based on each pixel position, the length, and the number of preset offset points; determine a first sub-offset based on the length and the number of preset offset points; and determine, based on the first scaling ratio and the first sub-offset, that each pixel position is located at the first pixel position in the preset block in a non-offset direction.

[0357] In some embodiments of the present application, the size information includes the length and width of the current block;

[0358] The second determination unit 20 is further configured to, when the offset direction is a vertical direction, determine, based on each pixel position, the width, and the number of preset offset points, a second pixel position in the preset block in a non-offset direction at each pixel position, wherein the non-offset direction is perpendicular to the offset direction; determine, based on the curve shape index, second curve information corresponding to the second pixel position; determine a second movement ratio based on the length and the number of preset offset points; and determine an offset amount based on the second curve information, the translation amplitude, and the second movement ratio, wherein the offset amount does not exceed the length of the current block.

[0359] In some embodiments of the present application, the second determination unit 20 is further configured to determine a second scaling ratio based on each pixel position, the width, and the number of preset offset points; determine a second sub-offset based on the width and the number of preset offset points; and determine, based on the second scaling ratio and the second sub-offset, that each pixel position is located in the second pixel position in the preset block in a non-offset direction.

[0360] In some embodiments of the present application, the curve division information is a division method in curve division;

[0361] The curve division is to translate the pixel points on the direct division line in the preset straight line geometric division to obtain a continuous curve division line to achieve geometric division.

[0362] In some embodiments of the present application, each pixel point of the straight dividing line is continuously translated according to different moving distances and the same moving direction to obtain the curved dividing line; or,

[0363] Each pixel point of the straight line dividing line is continuously translated according to different movement vectors to obtain the curved line dividing line.

[0364] In some embodiments of the present application, the movement distance is the distance between the corresponding pixel position on the preset curve shape and the straight dividing line;

[0365] The movement vector is a preset vector corresponding to a pixel position on a preset curve shape.

[0366] In some embodiments of the present application, the moving direction is formed by a combination of a horizontal direction and a vertical direction; or,

[0367] The moving direction is related to the division direction of the GPM.

[0368] In some embodiments of the present application, when the moving distance is not an integer pixel, the sub-pixel position is interpolated or rounded; or,

[0369] The moving distance is positively correlated with the length of the current block; or,

[0370] The moving distance is positively correlated with the smallest value between the length and width of the current block.

[0371] It is understandable that the encoder predicts the current block by using curve division in the GPM mode, so that the divided boundary is more closely aligned with the edge of the object or the edge of two parts with different motion vectors, thereby improving the encoding accuracy.

[0372] In practical applications, such as Figure 12As shown, an embodiment of the present application also provides an encoder, including: a second processor 23 and a second memory 24 storing executable instructions of the second processor 23, the second memory 24 relies on the second processor 23 to perform operations through a communication bus 25, and when the executable instructions are executed by the second processor 23, the inter-frame prediction method on the above-mentioned encoder side is executed.

[0373] In which, the second processor can be implemented by software, hardware, firmware or a combination thereof, and can use circuits, single or multiple dedicated integrated circuits, single or multiple general-purpose integrated circuits, single or multiple microprocessors, single or multiple programmable logic devices, or a combination of the aforementioned circuits or devices, or other suitable circuits or devices, so that the first processor can execute the corresponding steps of the inter-frame prediction method on the decoder side in the aforementioned embodiment.

[0374] An embodiment of the present application provides a computer-readable storage medium storing executable instructions. When the executable instructions are executed by one or more second processors, the second processors execute the inter-frame prediction method corresponding to the encoder.

[0375] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.

[0376] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0377] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0378] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0379] The above description is merely a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application.

Claims

1. An inter-frame prediction method, characterized in that: Applicable to decoders, including: Parse the code stream and determine the prediction mode parameters of the current block; When the prediction mode parameter indicates that an inter-frame geometric partitioning prediction mode GPM is used to determine a prediction value of the current block, determining a GPM mode parameter and curve partitioning information of the current block, wherein the GPM mode parameter includes angle index information and motion information of the current block; Determining a first reference block and a second reference block according to motion information of the current block; Determining, according to the angle index information, the curve division information, and the size information of the current block, a first weight value corresponding to a current point in the current block in the first reference block and a second weight value corresponding to the current point in the second reference block; Determine the prediction value of the current point in the current block according to the first prediction value of the current point in the first reference block, the first weight value, the second prediction value of the current point in the second reference block, and the second weight value.

2. The method according to claim 1, characterized in that The determining, according to the angle index information, the curve division information, and the size information of the current block, a first weight value corresponding to the current point in the current block in the first reference block and a second weight value corresponding to the current point in the second reference block, includes: determining an offset direction according to the angle index information; determining an offset according to the size information of the current block and the curve division information; According to the offset direction, the offset amount and the position information of the current point, a first weight value corresponding to the current point in the first reference block and a second weight value corresponding to the current point in the second reference block are determined.

3. The method according to claim 2, characterized in that The determining the offset direction according to the angle index information includes: When the angle index information indicates that the angle used by the GPM satisfies a horizontal determination condition, determining that the offset direction is a horizontal direction; When the angle index information indicates that the angle used by the GPM satisfies a vertical determination condition, the offset direction is determined to be a vertical direction.

4. The method according to claim 1, wherein The motion information of the current block includes first motion information and second motion information. The first reference block is determined according to the first motion information, and the second reference block is determined according to the second motion information. The first motion information is determined according to the first index information of the current block, and the second motion information is determined according to the second index information of the current block. The first index information and the second index information are obtained by parsing the code stream.

5. An inter-frame prediction method, characterized in that: Applications in encoders include: Determining prediction mode parameters for the current block; When the prediction mode parameter indicates that an inter-frame geometric partitioning prediction mode GPM is used to determine a prediction value of the current block, determining a GPM mode parameter and curve partitioning information of the current block, wherein the GPM mode parameter includes angle index information and motion information of the current block; Determining a first reference block and a second reference block according to motion information of the current block; Determining, according to the angle index information, the curve division information, and the size information of the current block, a first weight value corresponding to a current point in the current block in the first reference block and a second weight value corresponding to the current point in the second reference block; Determine the prediction value of the current point in the current block according to the first prediction value of the current point in the first reference block, the first weight value, the second prediction value of the current point in the second reference block, and the second weight value.

6. The method according to claim 5, characterized in that The determining, according to the angle index information, the curve division information, and the size information of the current block, a first weight value corresponding to the current point in the current block in the first reference block and a second weight value corresponding to the current point in the second reference block, includes: determining an offset direction according to the angle index information; determining an offset according to the size information of the current block and the curve division information; According to the offset direction, the offset amount and the position information of the current point, a first weight value corresponding to the current point in the first reference block and a second weight value corresponding to the current point in the second reference block are determined.

7. The method according to claim 6, characterized in that The determining the offset direction according to the angle index information includes: When the angle index information indicates that the angle used by the GPM satisfies a horizontal determination condition, determining that the offset direction is a horizontal direction; When the angle index information indicates that the angle used by the GPM satisfies a vertical determination condition, the offset direction is determined to be a vertical direction.

8. The method according to claim 5, characterized in that The motion information of the current block includes first motion information and second motion information, the first reference block is determined according to the first motion information, and the second reference block is determined according to the second motion information. The method further includes: encoding first index information corresponding to the first motion information and second index information corresponding to the second motion information, and writing the obtained encoding bits into a bitstream.

9. A decoder, characterized in that: include: A first processor and a first memory storing executable instructions of the first processor, wherein the first memory relies on the first processor to perform operations through a communication bus, and when the executable instructions are executed by the first processor, the inter-frame prediction method described in any one of claims 1 to 4 is executed.

10. An encoder, characterized in that include: A second processor and a second memory storing executable instructions of the second processor, the second memory relying on the second processor to perform operations through a communication bus, when the executable instructions are executed by the second processor, the inter-frame prediction method described in any one of claims 5 to 8 is executed.

11. A computer-readable storage medium, wherein: The computer-readable storage medium stores a code stream generated by a processor executing the inter-frame prediction method according to any one of claims 5 to 8.

Images