CN113473129B - An encoding and decoding method and device - Google Patents

Abstract

The present application provides an encoding and decoding method and device. The encoding method includes: obtaining a residual coefficient matrix corresponding to the current block; performing an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix; if the current block supports secondary Transformation, the current block supports the optional secondary transform AST technology, and the current block meets the first preset condition, then determine whether the current block meets the second preset condition; if the current block meets the second preset condition, then according to the The rate-distortion cost value corresponding to the current block determines whether to perform secondary transformation on the current block; wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second prediction The assumed condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction filtering mode. Improve encoding performance with this application.

Description

An encoding and decoding method and device

technical field

The present application relates to the technical field of encoding and decoding, and in particular, relates to an encoding and decoding method and device.

Background technique

In order to achieve the purpose of saving space, video images are transmitted after encoding, and a complete video encoding method may include processes such as prediction, transformation, quantization, entropy encoding, and filtering. Predictive coding may include intra coding and inter coding. Inter-frame coding is to use the correlation of the video time domain to predict the pixels of the current image by using the pixels of the adjacent coded image, so as to achieve the purpose of removing the redundancy of the video time domain. Intra-frame coding takes into account the strong spatial correlation between adjacent blocks, uses the surrounding reconstructed pixels as reference pixels to predict the current uncoded block, and only needs to perform subsequent coding processing on the residual value, while Instead of encoding the original value, the redundancy in the airspace is effectively removed, and the compression efficiency is greatly improved.

In the video coding process, transformation refers to converting an image described in the form of pixels in the spatial domain to an image in the transform domain, and expressing it in the form of transform coefficients. Since most images contain more flat areas and slowly changing areas, an appropriate transformation process can transform the scattered distribution of image energy in the spatial domain into a relatively concentrated distribution in the transformed domain, thereby eliminating signal differences. The frequency domain correlation among them, together with the quantization process, can effectively compress the code stream.

For the transformation process, a secondary transformation technique is proposed in the related art. The secondary transformation technique refers to: performing an initial transformation first to obtain transformation coefficients after the initial transformation. Then, the transformation coefficients after the initial transformation are transformed twice to obtain new transformation coefficients. Then, processes such as quantization and entropy coding are performed on the new transform coefficients. However, in practical applications, after the initial transformation, if the energy of the transformation coefficients after the initial transformation is concentrated enough, then when the transformation coefficients after the initial transformation are transformed again, overfitting may occur, thus Leading to problems such as poor encoding performance.

Contents of the invention

This application provides an encoding method, which is applied to the encoding end, and the method includes:

Obtain the residual coefficient matrix corresponding to the current block;

performing an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix;

If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the first preset condition, then determine whether the current block satisfies the second preset condition; if the current block satisfies the second preset condition condition, then determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block;

Wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

This application provides a decoding method, which is applied to the decoding end, and the method includes:

Get the encoded bitstream of the current block;

If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block meets the third preset condition, then determine whether the current block meets the fourth preset condition; if the current block meets the fourth preset condition condition, then determine whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix;

Wherein, the third preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the fourth preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

The present application provides an encoding device, which is applied to the encoding end, and the device includes:

An acquisition module, configured to acquire a residual coefficient matrix corresponding to the current block;

A transformation module, configured to perform an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix;

A determining module, configured to determine whether the current block satisfies the second preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the first preset condition; if the current block If the second preset condition is met, determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block;

Wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

The present application provides a decoding device, which is applied to the decoding end, and the device includes:

An acquisition module, configured to acquire the coded bitstream of the current block;

A determining module, configured to determine whether the current block satisfies the fourth preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the third preset condition; if the current block If the fourth preset condition is met, then it is determined whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix; wherein, the third preset condition includes: the current block The prediction mode of the block is an intra prediction mode; the fourth preset condition includes: the current block uses an intra prediction filtering mode, or does not use an intra prediction filtering mode.

The present application provides an encoding device, including: a processor and a machine-readable storage medium, where the machine-readable storage medium stores machine-executable instructions that can be executed by the processor;

The processor is used to execute machine-executable instructions to implement the following steps:

An acquisition module, configured to acquire the coded bitstream of the current block;

A determining module, configured to determine whether the current block satisfies the fourth preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the third preset condition; if the current block If the fourth preset condition is met, then it is determined whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix; wherein, the third preset condition includes: the current block The prediction mode of the block is an intra prediction mode; the fourth preset condition includes: the current block uses an intra prediction filtering mode, or does not use an intra prediction filtering mode.

The present application provides a decoding device, including: a processor and a machine-readable storage medium, where the machine-readable storage medium stores machine-executable instructions that can be executed by the processor;

The processor is used to execute machine-executable instructions to implement the following steps:

Get the encoded bitstream of the current block;

If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block meets the third preset condition, then determine whether the current block meets the fourth preset condition; if the current block meets the fourth preset condition condition, then determine whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix;

Wherein, the third preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the fourth preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

It can be seen from the above technical solutions that in the embodiment of the present application, it can be judged whether to allow secondary transformation on the current block, so as to eliminate blocks that do not need to undergo secondary transformation, and play a role in removing redundancy, thereby improving coding performance.

Description of drawings

1A-1C are schematic diagrams of a DT division mode in an embodiment of the present application;

FIG. 1D is a schematic diagram of an intra prediction mode in an embodiment of the present application;

FIG. 1E is a schematic diagram of a video coding framework in an implementation manner of the present application;

FIG. 2A is a flowchart of an encoding method in an embodiment of the present application;

FIG. 2B is a flowchart of a decoding method in an embodiment of the present application;

FIG. 3A is a flowchart of an encoding method in an embodiment of the present application;

FIG. 3B is a flowchart of a decoding method in an embodiment of the present application;

FIG. 4A is a flowchart of an encoding method in an embodiment of the present application;

FIG. 4B is a flowchart of a decoding method in an embodiment of the present application;

FIG. 5A is a structural diagram of an encoding device in an embodiment of the present application;

FIG. 5B is a structural diagram of a decoding device in an embodiment of the present application;

FIG. 6A is a hardware structural diagram of an encoding end device in an embodiment of the present application;

FIG. 6B is a hardware structural diagram of a decoder device in an implementation manner of the present application.

detailed description

The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not used to limit the present application. The singular forms of "a", "said" and "the" used in the embodiments and claims of this application are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein is meant to include any and all possible combinations of one or more of the associated listed items. It should be understood that although the embodiments of the present application may use terms such as first, second, and third to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of the present application, first information may also be called second information, and similarly, second information may also be called first information. Depending on the context, furthermore, the use of the word "if" could be interpreted as "at", or "when", or "in response to a determination".

In order to enable those skilled in the art to better understand the technical solutions of the present application, the technical terms are briefly explained below.

Prediction Signal: Refers to the pixel value derived from the encoded and decoded pixels. The residual is obtained by the difference between the original pixel and the predicted pixel, and then the process of residual transformation, quantization, and coefficient coding is performed.

Intra prediction (Intra Prediction): refers to the prediction mode that predicts the prediction value of the current block through the reconstructed pixel values of the surrounding decoded area. Common methods include copying pixels according to the angular direction, and deriving predicted values according to a certain gradient principle. Exemplarily, the intra-frame prediction removes the spatial domain correlation of the video/image.

Transform kernel: In video coding, transformation is an essential stage for data compression. Transformation can make the energy of the signal more concentrated, and based on discrete cosine transform (DCT, Discrete CosineTransform)/discrete sine transform (DST, Discrete Sine Transform) transformation technology has always been the mainstream transformation technology of video coding. DCT and DST are specifically divided into multiple transformation kernels according to different basis functions, as shown in Table 1, three kinds of transformation kernels are shown.

Table 1

Forward transform and inverse transform: In the video coding process, the process of forward transform and inverse transform is included. Forward transform is also called forward transform, and inverse transform is also called inverse transform. Exemplarily, the forward transform is to convert a two-dimensional residual coefficient (also called a residual signal) into a two-dimensional transform coefficient (also called a spectral signal) with more concentrated energy, and the transform coefficient is then subjected to processes such as quantization , it can effectively remove high-frequency components, retain low-frequency components, and play a role in compression. For example, the forward transformation process can be represented by the matrix form of formula (1).

F＝B·f· ^AT formula (1)

Exemplarily, M represents the width of the residual block (that is, a block composed of residual coefficients), N represents the height of the residual block, f represents the original residual signal of N*M dimension, and F represents the frequency domain of N*M dimension Signal (that is, the frequency domain signal after conversion), A and B represent M*M and N*N dimensional transformation matrices, A and B both satisfy orthogonality, as shown in Table 1 Transform kernel.

The inverse transform can be the reverse process of the forward transform, that is, the frequency domain signal F can be converted into the time domain residual signal f through the transformation matrix A and the transformation matrix B. For example, the inverse transformation process can be represented by the matrix form of formula (2).

f＝B ^T F A Formula (2)

Horizontal transform (Horizental transform) and vertical transform (Vertical transform): In the coding transformation stage, the input is a two-dimensional residual signal, assuming X=A·f ^T , then F=B·X ^T , for example, through the formula (3 ) in matrix form.

F＝B·f· ^AT ＝B·(A·f ^T ) ^T formula (3)

To sum up, it can be seen that the forward transformation of the two-dimensional signal can be realized by two one-dimensional forward transformation methods. After the first forward transformation, an M*N signal X is obtained, and the horizontal pixels of the two-dimensional residual signal are removed. Therefore, the first positive transformation is called the horizontal transformation, and A is called the horizontal transformation matrix. After the second forward transformation, the signal F is obtained, and the correlation between the pixels in the vertical direction of the two-dimensional residual signal is removed. Therefore, the second forward transformation is called vertical transformation, and B is called the vertical transformation matrix.

Transform pair (Transform pair): also known as transformation check. In order to support matrix blocks, M is not necessarily equal to N, and the dimensions of A and B are not necessarily equal. It can also support transformation matrices that A and B are not produced by the same transformation kernel, so , in the transformation, there is a transformation pair {H, V} composed of transformation kernels corresponding to A and B, H is called the horizontal transformation kernel, and V is called the vertical transformation kernel.

Rate-Distortion Principle (RDO, Rate-Distortion Optimized): There are two indicators for evaluating coding efficiency: code rate and PSNR (Peak Signal to Noise Ratio, peak signal-to-noise ratio). The smaller the bit stream, the greater the compression rate; the greater the PSNR, the better the quality of the reconstructed image. In mode selection, the discriminant formula is essentially a comprehensive evaluation of the two.

The cost value corresponding to the mode: J(mode)=D+λ*R. Exemplarily, D represents Distortion, which is usually measured by the SSE index. SSE refers to the mean square sum of the difference between the reconstruction block and the source image; λ is the Lagrangian multiplier; R is required for image block encoding in this mode The actual number of bits, including the sum of bits required for encoding mode information, motion information, residuals, etc. In mode selection, if the RDO principle is used to make comparison decisions on encoding modes, the best encoding performance can usually be guaranteed.

Implicit Transform Kernel Selection (IST, Implicit Selection of Transform): For intra-frame prediction residual blocks, the DCT2 transformation method is usually used, however, for most intra-frame prediction residual blocks, especially small blocks, DST7 performs better Therefore, the multi-transform core selection based on the coding unit (Code Unit, CU for short)-level (level) will further improve the coding performance. Based on this, a transform kernel selection method based on implicit expression is proposed, that is, the flag bit of the transform kernel (which may be called the IST flag bit) is hidden, so as to implicitly indicate the transform kernel used for transforming the current block.

Exemplarily, the encoder needs to use RDO to select whether the transform check is to use (DCT2, DCT2) or (DST7, DST7). In order to hide the IST flag, it is considered that when the transform check is selected as (DCT2, DCT2), the current block The number of non-zero transform coefficients is an even number. If the actual number of non-zero transform coefficients is an odd number, the encoding end sets the last non-zero transform coefficient to zero, so that the number of non-zero transform coefficients of the current block must be even. Similarly, when the transform check is selected as (DST7, DST7), the number of non-zero transform coefficients of the current block is an odd number, and if the actual number of non-zero transform coefficients is an even number, the encoder sets the last non-zero transform coefficient to The zero method makes the number of non-zero transform coefficients of the current block must be an odd number.

From the perspective of hardware implementation, the hardware implementation of the DST7 transform core requires more multiplication times, and the cost of hardware implementation is relatively high. Therefore, in order to reduce the complexity of hardware implementation, the DST7 transformation kernel may not be used for transformation blocks whose transformation coefficients exceed a certain range. The specific method is as follows: sr_x represents the abscissa of the rightmost non-zero transformation coefficient in the current transformation coefficient matrix, sr_y Represents the ordinate of the bottom non-zero transform coefficient in the current transform coefficient matrix. The area corresponding to all non-zero transform coefficients in the current transform coefficient matrix can be determined by sr_x and sr_y. The transform coefficients of the area beyond this range are all 0.

Since sr_x and sr_y need to be transmitted to the decoder, the decoder can directly obtain sr_x and sr_y. When sr_x is greater than or equal to 16 or sr_y is greater than or equal to 16, the decoder can directly determine that the transformation kernel used by the IST is (DCT2, DCT2). Then count the parity of the number of non-zero transform coefficients; otherwise, it is necessary to determine the parity of the number of non-zero transform coefficients of the current block, if the parity is odd, the transform kernel used by IST is (DST7, DST7), If the parity is even, the transformation kernel used by IST is (DCT2, DCT2), and the above process can be expressed by the following formula:

IstTuFlag=(sr_x>=16||sr_y>=16)? 0:(num_nz%2?1:0)

Based on the above analysis, it can be concluded that finally, the selection method of IST transformation kernel can be seen in Table 2:

Table 2

Exemplarily, the IST can be applied to the current block meeting the following conditions: the current block is in intra prediction mode; the division mode of the current sub-block is non-DT mode; the width of the current block is less than 64; the height of the current block is less than 64.

Derive Tree (DT for short) partitioning mode: DT partitioning mode is a new partitioning mode that generates new partitioning shapes and can further obtain performance gains. Exemplarily, the DT division mode may include a horizontal derivation mode and a vertical derivation mode, the schematic diagram of which may be as shown in FIG. 1A . The DT division mode can also be grown on the leaf nodes of the quadtree or the binary tree, as shown in FIG. 1B , for an I frame or a non-I frame, the DT division mode can obtain different PU divisions of the derivative mode by merging and dividing the boundaries of the CU ( 2N*hN, 2N*nU, 2N*nD, hN*2N, nL*2N or nR×2N).

Exemplarily, for the intra prediction mode, the current block predicted using the derivative mode can be transformed and quantized using 4 non-blocks without introducing a new transform kernel, and its schematic diagram can be shown in FIG. 1C .

Exemplarily, the current block using the horizontal derivation mode needs to meet the following two conditions: the height is greater than or equal to 16 and less than or equal to 64; the ratio of width to height is less than 4. Exemplarily, the current block using the vertical derivative mode needs to meet the following two conditions: the width is greater than or equal to 16 and less than or equal to 64; the ratio of height to width is less than 4.

Intra-frame prediction mode: In order to support finer angle prediction, the angle prediction mode can be extended to 62, plus 3 special intra-frame modes, a total of 65 modes. Referring to FIG. 1D , which is a schematic diagram of the intra prediction mode, the mode numbers 1-33 of the original angle prediction mode remain unchanged, and the mode number of the newly added angle prediction mode increases from 34 to 65.

Video encoding framework: Refer to Figure 1E, the video encoding framework can be used to implement the encoding end processing flow in the embodiment of this application, the schematic diagram of the video decoding framework is similar to Figure 1E, and will not be repeated here, and the video decoding framework can be used to implement this The processing flow of the decoding end in the application embodiment. In the video encoding framework and video decoding framework, it may include but not limited to: intra prediction, motion estimation/motion compensation, reference image buffer, in-loop filtering, reconstruction, transformation, quantization, inverse transformation, inverse quantization, entropy encoder and other modules. At the encoding end, through the cooperation between these modules, the processing flow at the encoding end can be realized, and at the decoding end, through the cooperation between these modules, the processing flow at the decoding end can be realized.

Exemplarily, in a video coding process, transformation refers to transforming an image described in the form of pixels in the spatial domain into an image in the transform domain, and expressing it in the form of transform coefficients. Since most images contain more flat areas and slowly changing areas, an appropriate transformation process can transform the scattered distribution of image energy in the spatial domain into a relatively concentrated distribution in the transformed domain, thereby eliminating signal differences. The frequency domain correlation among them, together with the quantization process, can effectively compress the code stream.

Exemplarily, entropy coding refers to the way of lossless coding according to the principle of information entropy, which is the last processing module of video compression, and is used to convert a series of element symbols used to represent video sequences into a code for transmission or storage The input symbols may include quantized transform coefficients, motion vector information, prediction mode information, transform and quantization related syntax, etc. The output data of the entropy coding module is the final code stream after the original video compression. Entropy coding can effectively remove the statistical redundancy of these video element symbols, and is one of the important tools to ensure the compression efficiency of video coding.

Secondary transformation: In the related art, different secondary transformation matrices are used for 4*4 blocks and non-4*4 blocks. From the encoding side, the transformation coefficient is obtained after (DCT2, DCT2) transformation, and then the 4*4 transformation coefficient in the upper left corner is multiplied by the secondary transformation matrix to obtain a new transformation coefficient, and then quantized and Entropy coding. From the perspective of the decoder, after inverse quantization of the transformation coefficients obtained from entropy decoding of the code stream, a second inverse transformation is performed on the 4*4 area in the upper left corner to obtain new transformation coefficients, and then the entire transformation block is used (DCT2 , DCT2) for inverse transformation to obtain residual coefficients.

Alternative secondary transform (AST, alternative secondary transform): For the intra-frame prediction block, the transformation coefficient obtained after the initial transformation can be selected to perform secondary transformation or not to perform secondary transformation, and the secondary transformation The purpose is to perform a second transformation on the low-frequency region, which further concentrates the energy. However, if the energy of the current block is concentrated enough, there is no need for secondary transformation. Therefore, based on the actual coding block, RDO can be used to determine whether to perform secondary transformation, and it can be expressed in an explicit or implicit way whether to perform secondary transformation transform. If the explicit AST is used, the secondary transformation indication information is explicitly encoded through the code stream, and if the implicit AST is used, the transformation coefficients are adjusted to meet the preset parity characteristics.

In related technologies, by default, the luma intra prediction residual block needs to be transformed twice, but after the initial transformation, if the energy of the transformation coefficient after the initial transformation is concentrated enough, then the transformation coefficient after the initial transformation is transformed twice , there may be over-fitting phenomenon, leading to problems such as poor coding performance, that is, the coding effect of the secondary transformation is not ideal.

In view of the above findings, in the embodiment of the present application, the encoding end can judge whether to allow the second transformation of the current block according to the first preset condition and the second preset condition, if the current block satisfies the first preset condition, and the current block also If the second preset condition is met, the current block is allowed to undergo secondary transformation. Based on this, it may be determined whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block does not meet the first preset condition, it is prohibited to perform secondary transformation on the current block, so as to eliminate blocks that do not need to undergo secondary transformation, play a role in removing redundancy, and improve coding performance. If the current block satisfies the first preset condition and the current block does not satisfy the second preset condition, it is directly determined to perform secondary transformation on the current block.

In a possible implementation manner, for the encoding end, RDO can be used to determine whether to perform secondary transformation on the current block, and explicitly encode indication information (such as secTuflag) to indicate whether to perform secondary transformation on the current block, For example, when the value of secTuflag is the first value, it may indicate that a second transformation is performed on the current block; when the value of secTuflag is the second value, it may indicate that the current block is not to be transformed twice. For the decoding end, after obtaining the coded bit stream, the secTuflag of the current block can be parsed from the coded bit stream. If the value of secTuflag is the first value, it is determined to perform a second inverse transform on the current block. If If the value of secTuflag is the second value, it is determined not to perform the second inverse transform on the current block.

In another possible implementation, for the encoding end, RDO can be used to determine whether to perform secondary transformation on the current block, and the target characteristics (such as parity, etc.) of the target transformation coefficient matrix can be used to implicitly indicate whether Perform secondary transformation on the current block, instead of explicitly indicating whether to perform secondary transformation on the current block through indication information, for example, implicitly express secTuflag through the parity of the number of odd transformation coefficients in the target transformation coefficient matrix , to determine whether the current block uses secondary transformation. For the decoder, after obtaining the encoded bit stream, the target transform coefficient matrix is parsed from the encoded bit stream, and the target value of the preset flag bit of the current block is determined according to the target characteristics of the target transform coefficient matrix, that is, the derivation Output the target value of the preset flag bit of the current block. If the target value of the preset flag bit is the first value, it is determined to perform a second inverse transform on the current block. If the target value of the preset flag bit is the second value, it is determined not to perform the second inverse transform on the current block.

Exemplarily, based on the above two situations, for the encoding side, if it is determined that the current block needs to be re-transformed, it can also decide to perform a re-transform on the 4*4 area in the upper left corner based on the size constraint condition, or 8 *8 area undergoes secondary transformation. For the decoder, if it is determined that a second inverse transformation is required for the current block, it can also decide whether to perform a second inverse transformation on the 4*4 area in the upper left corner or an 8*8 area based on the size constraints .

The encoding method and the decoding method in the embodiment of the present application will be described in detail below in conjunction with several specific embodiments.

Embodiment 1: Referring to FIG. 2A, it is a schematic flow diagram of an encoding method, which can be applied to the encoding end. The method includes:

Step 211, acquire the residual coefficient matrix corresponding to the current block.

Step 212, performing an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix.

Step 213, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the first preset condition, then determine whether the current block satisfies the second preset condition; if the current block satisfies the second preset condition, then According to the rate-distortion cost value corresponding to the current block, it is determined whether to perform secondary transformation on the current block.

Exemplarily, the first preset condition includes that the prediction mode of the current block is an intra prediction mode. The second preset condition includes: the current block uses the intra-frame prediction filtering mode, or does not use the intra-frame prediction filtering mode.

In a possible implementation manner, the first preset condition may include at least one of the following conditions: the current block is a luma block; and the initial transform coefficient matrix of the current block contains non-zero transform coefficients.

In a possible implementation manner, the first preset condition may also include: the mode number of the intra prediction mode adopted by the current block is in the first mode number range, and the reference samples on the left outside the current block are available; or, the current block The mode number of the adopted intra-frame prediction mode is located in the second mode number interval, and the reference samples on the outside of the current block are available.

Exemplarily, the first range of mode numbers may include but not limited to: mode number 0 to mode number 2, mode number 13 to mode number 32, and mode number 44 to mode number 65. The second range of mode numbers may include but not limited to: mode number 0 to mode number 23, and mode number 34 to mode number 57. Of course, the above are only examples of the first mode number interval and the second mode number interval, and there is no limitation on this.

In a possible implementation manner, the first preset condition may further include: the current block adopts the transformation checker (DCT2, DCT2) for initial transformation; or, the current block does not use the transformation kernel (DST7, DST7) for initial transformation.

Exemplarily, in the IST technology, see Table 2, if istTuflag is equal to the first value (such as 0), it means that the current block uses the transformation check (DCT2, DCT2) for initial transformation, if istTuflag is equal to the second value ( As in 1), it means that the current block uses the transformation checker (DST7, DST7) for initial transformation. Based on this, the first preset condition may further include: istTuflag=0, that is, istTuflag=0 indicates that the previous block uses the transform check (DCT2, DCT2) for initial transformation.

In a possible implementation manner, the second preset condition may also include one of the following conditions:

The width and height of the current block satisfy a first size condition; or,

The current block supports DT division mode; or,

The current block does not support the DT division mode; or,

The width and height of the current block meet the first size condition, and the current block supports the DT division mode; or,

The width and height of the current block satisfy the first size condition, and the current block does not support the DT division mode.

Exemplarily, if the width of the current block is less than or equal to the first value, and the height of the current block is less than or equal to the second value, then the width and height of the current block satisfy the first size condition. For example, the first value may be 32 and the second value may be 32, or the first value may be 64 and the second value may be 64. Of course, the above is just an example, without limitation.

Exemplarily, after the encoding end determines whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block, if the current block is subjected to secondary transformation, the specified area in the upper left corner of the initial transformation coefficient matrix The transform coefficients are transformed twice to obtain the target transform coefficient matrix; if the current block is not transformed twice, the initial transform coefficient matrix is determined as the target transform coefficient matrix. Encoding is performed according to the target transform coefficient matrix to obtain an encoded bit stream of the current block.

In a possible implementation manner, the encoding end may explicitly encode indication information, and the indication information indicates whether to perform secondary transformation on the current block. Based on this, the encoder can directly encode the target transformation coefficient matrix in the encoded bitstream. On this basis, if the current block is transformed twice, the first indication information can be added to the encoded bitstream. The first indication information It is used to indicate the secondary transformation of the current block. If secondary transformation is not performed on the current block, second indication information may be added to the coded bit stream, where the second indication information is used to indicate that secondary transformation is not performed on the current block.

In another possible implementation manner, whether to perform secondary transformation on the current block may be implicitly indicated by the target feature (such as parity) of the target transformation coefficient matrix, instead of explicitly indicating whether to perform the second transformation on the current block through indication information Perform a secondary transformation. Based on this, encoding is performed according to the target transform coefficient matrix to obtain the encoded bit stream of the current block, including: determining the target feature of the target transform coefficient matrix according to the transform coefficients in the target transform coefficient matrix; according to the target feature and the preset of the current block The target value of the flag bit determines whether to adjust the target transformation coefficient matrix; when the target value of the preset flag bit is the first value, it is used to indicate the secondary transformation of the current block, and the target value of the preset flag bit When the value is the second value, it is used to indicate that no secondary transformation is performed on the current block. If the target transformation coefficient matrix is not adjusted, the target transformation coefficient matrix is encoded to obtain the encoded bit stream of the current block; if the target transformation coefficient matrix is adjusted, the transformation coefficients in the target transformation coefficient matrix are adjusted so that The target feature of the adjusted target transform coefficient matrix matches the target value of the preset flag bit, and the adjusted target transform coefficient matrix is encoded to obtain an encoded bit stream of the current block.

It can be seen from the above technical solutions that in the embodiment of the present application, the encoding end can judge whether to allow the second transformation of the current block according to the first preset condition and the second preset condition. If the current block satisfies the first preset condition, and the current If the block also satisfies the second preset condition, the current block is allowed to undergo secondary transformation. Based on this, it may be determined whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block does not meet the first preset condition, it is prohibited to perform secondary transformation on the current block, so as to eliminate blocks that do not need to undergo secondary transformation, play a role in removing redundancy, and improve coding performance.

Embodiment 2: Referring to FIG. 2B, it is a schematic flow diagram of a decoding method, which can be applied to a decoding end. The method includes:

Step 221, acquire the coded bit stream of the current block.

Step 222, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the third preset condition, then determine whether the current block satisfies the fourth preset condition; if the current block satisfies the fourth preset condition, then Whether to perform secondary inverse transformation on the current block is determined according to the indication information in the coded bit stream or the target transformation coefficient matrix.

The third preset condition may include that the prediction mode of the current block is an intra-frame prediction mode; the fourth preset condition may include: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction filtering mode.

In a possible implementation manner, the third preset condition may include at least one of the following conditions: the current block is a luma block; and the initial transform coefficient matrix of the current block contains non-zero transform coefficients.

In a possible implementation manner, the third preset condition may also include: the mode number of the intra prediction mode adopted by the current block is in the first mode number interval, and the reference samples on the left outside the current block are available; or, the current block The mode number of the adopted intra-frame prediction mode is located in the second mode number interval, and the reference samples on the outside of the current block are available.

Exemplarily, the first range of mode numbers may include but not limited to: mode number 0 to mode number 2, mode number 13 to mode number 32, and mode number 44 to mode number 65. The second range of mode numbers may include but not limited to: mode number 0 to mode number 23, and mode number 34 to mode number 57. Of course, the above are only examples of the first mode number interval and the second mode number interval, and there is no limitation on this.

In a possible implementation manner, the third preset condition may also include: the current block uses the transform check (DCT2, DCT2) to perform initial inverse transform; or, the current block does not use the transform check (DST7, DST7) to perform initial inverse transform .

Exemplarily, in the IST technology, if istTuflag is equal to the first value (such as 0), it means that the current block uses the transformation check (DCT2, DCT2) to perform initial inverse transformation; if istTuflag is equal to the second value (such as 1), it means The current block is initially inverse transformed using transform checks (DST7, DST7). Based on this, the third preset condition may further include: istTuflag=0, that is, istTuflag=0 indicates that the previous block uses the transform check (DCT2, DCT2) to perform initial inverse transform.

In a possible implementation manner, the fourth preset condition may also include one of the following conditions:

The width and height of the current block satisfy a first size condition; or,

The current block supports DT division mode; or,

The current block does not support the DT division mode; or,

The width and height of the current block meet the first size condition, and the current block supports the DT division mode; or,

The width and height of the current block satisfy the first size condition, and the current block does not support the DT division mode.

Exemplarily, if the width of the current block is less than or equal to the first value, and the height of the current block is less than or equal to the second value, then the width and height of the current block satisfy the first size condition. For example, the first value may be 32 and the second value may be 32, or the first value may be 64 and the second value may be 64. Of course, the above is just an example, without limitation.

In a possible implementation manner, the encoding end may explicitly encode indication information to indicate whether to perform secondary transformation on the current block. Based on this, the decoding end determines whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream, which may include: parsing out the indication information from the coded bit stream. If the indication information is the first indication information, determine to perform a second inverse transformation on the current block; the first indication information is used to indicate to perform a second transformation on the current block; or, if the indication information is the second indication information, determine No secondary inverse transformation is performed on the current block; the second indication information is used to indicate that no secondary transformation is performed on the current block.

In another possible implementation manner, the encoding end may implicitly indicate whether to perform secondary transformation on the current block through the target feature (such as parity) of the target transformation coefficient matrix, instead of explicitly indicating through indication information. Based on this, the decoder determines whether to perform a second inverse transform on the current block according to the target transform coefficient matrix in the coded bit stream, which may include: parsing out the target transform coefficient matrix from the coded bit stream; transform_coefficients, which identifies the target features for the target transform coefficient matrix. Determine the target value of the preset flag bit of the current block according to the target feature; when the target value of the preset flag bit is the first value, it is used to indicate that the current block is subjected to secondary inverse transformation, and the target value of the preset flag bit is When the value is the second value, it is used to indicate that the second inverse transformation is not performed on the current block. Whether to perform secondary inverse transformation on the current block is determined according to the target value of the preset flag bit.

Exemplarily, after determining whether to perform a second inverse transformation on the current block according to the coded bit stream, the decoder can parse out the target transformation coefficient matrix from the coded bit stream; The transformation coefficients in the specified area in the upper left corner of the matrix are subjected to secondary inverse transformation to obtain the initial transformation coefficient matrix; if the current block is not subjected to secondary inverse transformation, the target transformation coefficient matrix is determined as the initial transformation coefficient matrix. The initial inverse transformation is performed on the initial transformation coefficient matrix to obtain the residual coefficient matrix corresponding to the current block, and the reconstruction value of the current block is determined according to the residual coefficient matrix.

It can be seen from the above technical solutions that in the embodiment of the present application, the decoding end can judge whether to allow the second inverse transformation of the current block according to the third preset condition and the fourth preset condition, for example, if the current block satisfies the third preset condition condition, and the current block also satisfies the fourth preset condition, then the current block is allowed to be subjected to secondary inverse transformation, and whether to perform secondary inverse transformation on the current block is determined according to the coded bit stream of the current block. If the current block does not meet the third preset condition, the second inverse transformation of the current block is prohibited, that is, the decoding end directly determines that the current block does not undergo the second inverse transformation, thereby eliminating the blocks that do not need to undergo the second transformation. The role of de-redundancy, thereby improving the coding performance.

Embodiment 3: Referring to FIG. 3A, it is a schematic flow diagram of an encoding method, which can be applied to the encoding end. The method includes:

Step 311, acquire a residual coefficient matrix corresponding to the current block.

Exemplarily, the reference block corresponding to the current block may be determined, and for each pixel point of the current block, the reference point corresponding to the pixel point is determined from the reference block, and the difference between the pixel value of the pixel point and the pixel value of the reference point The difference between them is the residual coefficient corresponding to the pixel, and the residual coefficients corresponding to all the pixels of the current block form the residual coefficient matrix corresponding to the current block.

Step 312, performing an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix.

Exemplarily, the transformation check (DCT2, DCT2) can be used to perform initial transformation on the residual coefficient matrix to obtain the initial transformation coefficient matrix, and the transformation correction (DST7, DST7) can also be used to perform initial transformation on the residual coefficient matrix to obtain the initial transformation coefficient matrix. Exemplarily, when the transformation check (DCT2, DCT2) is used to perform initial transformation on the residual coefficient matrix, the initial transformation process can be shown in formula (3), matrix A is DCT2 (see Table 1), and matrix B is also DCT2. When the transformation check (DST7, DST7) is used to perform initial transformation on the residual coefficient matrix, the initial transformation process can be shown in formula (3), matrix A is DST7 (see Table 1), and matrix B is also DST7.

Regarding the use of transformation check (DCT2, DCT2) for the initial transformation of the residual coefficient matrix, or the use of transformation check (DST7, DST7) for the initial transformation of the residual coefficient matrix, the following methods can be used: sr_x represents the rightmost part of the residual coefficient matrix The abscissa of the non-zero transformation coefficient of , sr_y represents the ordinate of the bottom non-zero transformation coefficient in the residual coefficient matrix, sr_x and sr_y can determine the area corresponding to all non-zero transformation coefficients in the residual coefficient matrix, and the area beyond this range Areas are all 0. If sr_x is greater than or equal to 16 or sr_y is greater than or equal to 16, the encoding end uses the transform check (DCT2, DCT2) to initially transform the residual coefficient matrix. Otherwise, the encoding end chooses to use transform check (DCT2, DCT2) or transform check (DST7, DST7) through RDO. For example, if the rate-distortion cost value of the transform check (DCT2, DCT2) is smaller than the rate-distortion cost value of the transform check (DST7, DST7), then the residual coefficient matrix is initially transformed using the transform check (DCT2, DCT2); if the transform check The rate-distortion cost value of (DCT2, DCT2) is greater than the rate-distortion cost value of the transformation check (DST7, DST7), and the transformation check (DST7, DST7) is used to initially transform the residual coefficient matrix.

Of course, the above is only an example of performing initial transformation on the residual coefficient matrix to obtain the initial transformation coefficient matrix, and there is no limitation on this, as long as the initial transformation can be performed on the residual coefficient matrix to obtain the initial transformation coefficient matrix.

Step 313, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the first preset condition, then determine whether the current block satisfies the second preset condition; if the current block satisfies the second preset condition, then According to the rate-distortion cost value corresponding to the current block, it is determined whether to perform secondary transformation on the current block.

Exemplarily, if the value of the secondary transformation enabling flag (such as SecondaryTransformEnableFlag) is a preset value (such as 1), it is determined that the current block supports secondary transformation. When the value of the secondary transformation enable flag is a preset value, it means that the secondary transformation is allowed to be enabled at the sequence level or the frame level, that is, the current block is allowed to enable the secondary transformation. Therefore, the encoder determines that the current block supports the secondary transformation.

Exemplarily, if the value of the optional secondary transform enabling flag (such as AstEnableFlag) is a preset value (such as 1), it is determined that the current block supports AST technology. When the optional secondary transform enabling flag is a preset value, it means that the sequence level or frame level allows enabling the AST technology, that is, allowing the current block to enable the AST technology, therefore, the encoder determines that the current block supports the AST technology.

Exemplarily, if the current block does not support secondary transformation, the encoder determines not to perform secondary transformation on the current block. If the current block supports the secondary transformation, but the current block does not support the AST technology, the encoder determines not to perform the secondary transformation on the current block.

If the current block supports secondary transformation, and the current block supports AST technology, the encoder determines whether the current block satisfies the first preset condition. If the current block does not meet the first preset condition, the encoder determines not to perform secondary transformation on the current block.

If the current block supports secondary transformation, and the current block supports AST technology, and the current block satisfies the first preset condition, then the encoder determines whether the current block satisfies the second preset condition. If the current block does not satisfy the second preset condition, the encoding end directly determines to perform the second transformation on the current block, instead of using RDO or other methods to decide whether to perform the second transformation.

If the current block supports secondary transformation, and the current block supports AST technology, and the current block satisfies the first preset condition, then the encoder determines whether the current block satisfies the second preset condition. If the current block satisfies the second preset condition, the encoder determines whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. For example, a first rate-distortion cost value is determined when only initial transformation is performed on the current block, and a second rate-distortion cost value is determined when initial transformation is performed on the current block and then secondary transformation is performed. If the second rate-distortion cost is smaller than the first rate-distortion cost, it is determined to perform a second transform on the current block. If the second rate-distortion cost is greater than the first rate-distortion cost, it is determined not to perform secondary transformation on the current block.

When determining the first rate-distortion cost value when only the initial transformation is performed on the current block, the encoder performs initial transformation on the residual coefficient matrix to obtain the initial transformation coefficient matrix, and determines the first rate-distortion corresponding to the current block based on the initial transformation coefficient matrix The cost value is not limited to the determination method, and RDO may be used to determine the first rate-distortion cost value corresponding to the current block.

When determining the second rate-distortion cost value when performing the initial transformation on the current block and then performing the secondary transformation, the encoding end performs the initial transformation on the residual coefficient matrix to obtain the initial transformation coefficient matrix, and performs a secondary transformation on the initial transformation coefficient matrix Transform to obtain the target transform coefficient matrix, and determine the second rate-distortion cost value corresponding to the current block based on the target transform coefficient matrix.

In the above embodiments, it is necessary to determine whether the current block satisfies the first preset condition. For the first preset condition, in a possible implementation manner, the first preset condition may include but not limited to at least one of the following conditions :

Condition 1: the prediction mode of the current block is an intra prediction mode.

Condition 2: the current block is a luma block.

Condition 3: The initial transform coefficient matrix of the current block contains non-zero transform coefficients.

The initial transformation coefficient matrix of the current block contains non-zero transformation coefficients, which means that after the initial transformation is performed on the residual coefficient matrix to obtain the initial transformation coefficient matrix, if all the transformation coefficients in the initial transformation coefficient matrix are zero, it means that the initial The transformation coefficient matrix does not contain non-zero transformation coefficients. In this case, the position of the cbf flag of the current block is the first value (such as the value 0), and the transformation coefficient of the current block does not need to be encoded. If all the transform coefficients in the initial transform coefficient matrix are not zero (that is, any one or more transform coefficients are not zero), it means that the initial transform coefficient matrix contains non-zero transform coefficients. In this case, the cbf of the current block The flag position is the second value (for example, value 1), which needs to encode the transform coefficient of the current block.

Condition 4: The mode number of the intra prediction mode adopted by the current block is located in the first mode number interval, and the reference samples on the left outside the current block are available; or, the mode number of the intra prediction mode adopted by the current block is located in the second mode number interval , and the reference samples above and outside the current block are available. Exemplarily, the first mode number interval may include mode number 0 to mode number 2, mode number 13 to mode number 32, and mode number 44 to mode number 65. The second mode number interval may include mode number 0 to mode number 23, and mode number 34 to mode number 57. Of course, the above are only examples of the first mode number interval and the second mode number interval, and there is no limitation on this.

Condition 5: The current block uses the transform check (DCT2, DCT2) for initial transform; or, the current block does not use the transform check (DST7, DST7) for initial transform. Alternatively, istTuflag is equal to the first value (such as 0).

Referring to step 312, the coding end uses the transform check (DCT2, DCT2) to perform an initial transform on the residual coefficient matrix, or uses the transform check (DST7, DST7) to perform an initial transform on the residual coefficient matrix.

If the transformation check (DCT2, DCT2) is used to initially transform the residual coefficient matrix, condition 5 is satisfied; if the transformation check (DST7, DST7) is used to initially transform the residual coefficient matrix, then condition 5 is not satisfied.

Exemplarily, the first preset condition may only include condition 1 and condition 2. Alternatively, the first preset condition may only include condition 1, condition 2 and condition 5. Alternatively, the first preset condition may include condition 1, condition 2, and condition 5. On this basis, the first preset condition may also include condition 4 and/or condition 5. Of course, the above is just an example and is not limited to this .

When the first preset condition includes condition 4, the first preset condition only includes one case of condition 4. For example, the first preset condition may include that the mode number of the intra prediction mode adopted by the current block is in the first mode number interval, and reference samples to the left outside the current block are available. Alternatively, the first preset condition may include that the mode number of the intra prediction mode adopted by the current block is in the second mode number interval, and the upper and outer reference samples of the current block are available.

When the first preset condition includes condition 5, the first preset condition only needs to include one condition of condition 5. For example, the first preset condition may include that the current block is initially transformed using transform check (DCT2, DCT2). Alternatively, the first preset condition may include that the current block does not use the transform check (DST7, DST7) for initial transformation.

The following describes different situations of the first preset condition in conjunction with several specific application scenarios.

Application scenario 1: the first preset condition may include: the prediction mode of the current block is an intra prediction mode.

Application scenario 2: the first preset condition may include: the prediction mode of the current block is an intra prediction mode. The current block is a luma block.

Application scenario 3: the first preset condition may include: the prediction mode of the current block is an intra prediction mode. The current block is a luma block. The current block is initially transformed using transform checks (DCT2, DCT2).

Application scenario 4: the first preset condition may include: the prediction mode of the current block is an intra prediction mode. The current block is a luma block. The current block was not initially transformed with transform checks (DST7, DST7).

Application scenario 5: On the basis of any application scenario in application scenarios 1-4, the first preset condition may also include: the mode number of the intra prediction mode adopted by the current block is located in the first mode number interval, and the current block is outside The reference sample on the left is available.

Application scenario 6: On the basis of any application scenario in application scenarios 1-4, the first preset condition may also include: the mode number of the intra prediction mode adopted by the current block is located in the second mode number interval, and the current block is outside The above reference samples are available.

Application scenario 7: On the basis of any application scenario in application scenarios 1-6, the first preset condition may further include: the initial transform coefficient matrix of the current block contains non-zero transform coefficients.

Of course, the above application scenarios 1-7 are just a few examples of the first preset conditions, and there is no limitation on the first preset conditions. Taking application scenario 3 as an example, the current block meets the first preset condition means: if the prediction mode of the current block is the intra prediction mode, the current block is a luma block, and the current block is initially transformed using transform check (DCT2, DCT2), It means that the current block satisfies the first preset condition. If the prediction mode of the current block is not the intra prediction mode, it means that the current block does not meet the first preset condition, or if the current block is not a luma block, it means that the current block does not meet the first preset condition, or, if If the current block does not use the transform check (DCT2, DCT2) for initial transformation, it means that the current block does not satisfy the first preset condition.

In the above embodiment, it is necessary to judge whether the current block satisfies the second preset condition. For the second preset condition, in a possible implementation manner, the second preset condition may include but not limited to at least one of the following conditions :

Condition a: the current block uses the intra prediction filtering mode, or the current block does not use the intra prediction filtering mode.

Condition b: the current block supports the DT division mode; or, the current block does not support the DT division mode.

Condition c: the width and height of the current block satisfy the first size condition. For example, if the width of the current block is less than or equal to the first value, and the height of the current block is less than or equal to the second value, then the width and height of the current block satisfy the first size condition.

For example, the first numerical value can be 32 or 64, and the second numerical value can be 32 or 64. Of course, 32 and 64 are just examples, which are not limited. For example, the first numerical value can also be 128, and the second numerical value can also be 128. Wait. For example, the first numerical value is 32 and the second numerical value is 32, or the first numerical value may be 64 and the second numerical value may be 64.

Exemplarily, the second preset condition may only include condition a. Alternatively, the second preset condition may only include condition a and condition b. Alternatively, the second preset condition may only include condition a and condition c. Alternatively, the second preset condition may only include condition b. Alternatively, the second preset condition may only include condition b and condition c. Alternatively, the second preset condition may include condition a, condition b and condition c. Of course, the above is just an example, without limitation.

When the second preset condition includes condition a, it only needs to include one case of condition a. For example, the second preset condition includes that the current block uses an intra prediction filtering mode. Alternatively, the second preset condition includes that the current block does not use an intra prediction filtering mode.

When the second preset condition includes condition b, only one case of condition b may be included. For example, the second preset condition includes that the current block supports the DT division mode. Alternatively, the second preset condition includes that the current block does not support the DT division mode.

The following describes different situations of the second preset condition in conjunction with several specific application scenarios.

Application scenario 8: the second preset condition may include: the current block uses an intra prediction filtering mode.

Application scenario 9: the second preset condition may include: the current block does not use the intra prediction filtering mode.

Application Scenario 10: The second preset condition may include: the current block supports the DT division mode.

Application Scenario 11: The second preset condition may include: the current block does not support the DT division mode.

Application scenario 12: the second preset condition may include: the current block uses an intra prediction filtering mode. The current block supports DT partition mode.

Application scenario 13: the second preset condition may include: the current block uses an intra prediction filtering mode. The current block does not support DT partition mode.

Application scenario 14: the second preset condition may include: the current block does not use the intra prediction filtering mode. The current block supports DT partition mode.

Application scenario 15: the second preset condition may include: the current block does not use the intra prediction filtering mode. The current block does not support DT partition mode.

Application scenario 16: On the basis of any application scenario in application scenarios 8-15, the second preset condition may further include: the width and height of the current block satisfy the first size condition.

Of course, the above-mentioned application scenarios 8-16 are just a few examples of the second preset conditions, and there is no limitation on the second preset conditions. Taking application scenario 12 as an example, the fact that the current block satisfies the second preset condition means: if the current block uses the intra prediction filtering mode and the current block supports the DT division mode, it means that the current block meets the second preset condition. If the current block does not use the intra prediction filtering mode, it means that the current block does not meet the second preset condition, or if the current block does not support the DT division mode, it means that the current block does not meet the second preset condition.

Step 314: If the current block is to be transformed twice, then the transform coefficients in the specified area in the upper left corner of the initial transform coefficient matrix are transformed twice to obtain the target transform coefficient matrix; if the current block is not transformed twice, then The initial transform coefficient matrix is determined as the target transform coefficient matrix. So far, the target transformation coefficient matrix of the current block can be obtained.

In a possible implementation manner, performing a second transformation on the transformation coefficients of the specified area located in the upper left corner in the initial transformation coefficient matrix may include but not limited to: if the width and height of the current block meet the second size condition, based on The secondary transformation matrix of the first specified size performs secondary transformation on the transformation coefficients of the first specified area located in the upper left corner of the initial transformation coefficient matrix. If the width and height of the current block do not meet the second size condition, perform secondary transformation on the transformation coefficients of the second specified area located in the upper left corner in the initial transformation coefficient matrix based on the secondary transformation matrix of the second specified size.

Exemplarily, the size of the first specified area is the first specified size, the size of the second specified area is the second specified size, and the first specified size is larger than the second specified size. For example, the first specified size is 8*8; the second specified size is 4*4. Of course, the above are only examples of the first specified size and the second specified size, and there is no limitation to the first specified size and the second specified size.

The width and height of the current block meet the second size condition, which may include but not limited to: the width of the current block is greater than or equal to the third value, the height of the current block is greater than or equal to the fourth value; or, the width of the current block is greater than or equal to the fifth value, The height of the current block is greater than the sixth value; or, the width of the current block is greater than the seventh value, and the height of the current block is greater than or equal to the eighth value.

Exemplarily, the above-mentioned third value, fourth value, fifth value, sixth value, seventh value, and eighth value can all be configured according to experience, and these values are not limited. For example, the third numerical value may be 8 or 16, the fourth numerical value may be 8 or 16, the fifth numerical value may be 8 or 16, and the sixth numerical value may be 8 or 16. The seventh numerical value may be 8 or 16, and the eighth numerical value may be 8 or 16.

In summary, if the width and height of the current block meet the second size condition (such as the width of the current block is greater than or equal to 8, and the height of the current block is greater than or equal to 8), then based on the secondary transformation matrix with a size of 8*8, the The transformation coefficients of the first specified area (ie, the area with a size of 8*8) located in the upper left corner in the initial transformation coefficient matrix are subjected to secondary transformation. If the width and height of the current block do not meet the second size condition, then based on the secondary transformation matrix with a size of 4*4, the second specified area located in the upper left corner of the initial transformation coefficient matrix (that is, the area with a size of 4*4) ) transformation coefficients for secondary transformation.

Step 315, perform encoding according to the target transform coefficient matrix to obtain an encoded bit stream of the current block.

Step 316: If the current block is to be re-transformed, first indication information is added to the coded bit stream of the current block, and the first indication information is used to instruct the current block to be re-transformed. If the current block is not to be re-transformed, second indication information is added to the coded bit stream of the current block, and the second indication information is used to indicate that the current block is not to be re-transformed.

Exemplarily, the encoding end may explicitly encode indication information, and the indication information indicates whether to perform secondary transformation on the current block. Based on this, after obtaining the target transformation coefficient matrix, the encoder can directly encode the target transformation coefficient matrix in the coded bitstream of the current block, and add the first indication information or the second indication information in the coded bitstream of the current block.

Exemplarily, the target transformation coefficient matrix is transformed into coded bits and streamed to the decoder through the entropy coding process. In the process of encoding according to the target transformation coefficient matrix, if the current block uses the (DST7, DST7) transformation kernel for initial transformation, then after The target transformation coefficient matrix after transformation and quantization needs to meet: sr_x and sr_y are less than 16, and the number of all non-zero transformation coefficients in the target transformation coefficient matrix is an odd number, which is used to indicate that the current block uses (DST7, DST7) transformation kernel for initial transformation . Or, if the current block uses the (DCT2, DCT2) transform kernel for initial transformation, the target transform coefficient matrix after transformation and quantization needs to satisfy: sr_x or sr_y is greater than or equal to 16, or, all non-zero transform coefficients in the target transform coefficient matrix The number is an even number, which is used to indicate that the current block uses the (DCT2, DCT2) transform kernel for initial transformation.

Exemplarily, if the current block supports secondary transformation, the current block supports AST technology, the current block satisfies the first preset condition, and the current block satisfies the second preset condition, it is determined according to the rate-distortion cost value corresponding to the current block After performing secondary transformation, add indication information to the coded bitstream of the current block, for example, if the current block is subjected to secondary transformation, add the first indication information to the coded bitstream, if the current block is not subjected to secondary transformation, The second indication information is added to the coded bit stream.

If the current block does not support secondary transformation, it is determined that secondary transformation is not performed on the current block, but no indication information needs to be added to the coded bit stream of the current block. Alternatively, if the current block supports secondary transformation but the current block does not support AST technology, it is determined that secondary transformation is not performed on the current block, but no indication information needs to be added to the coded bit stream of the current block. Or, if the current block supports secondary transformation, the current block supports AST technology, and the current block does not meet the first preset condition, it is determined that the current block is not subjected to secondary transformation, but there is no need to add an indication to the coded bitstream of the current block information. Or, if the current block supports secondary transformation, the current block supports AST technology, the current block satisfies the first preset condition, and the current block does not meet the second preset condition, then it is determined to perform secondary transformation on the current block, but it is not necessary to Indication information is added to the coded bitstream of the block.

In the above case, it is not necessary to add indication information to the coded bit stream of the current block, which means that indication information for indicating that the current block does not undergo secondary transformation or that the current block undergoes secondary transformation does not need to be added.

It can be seen from the above technical solutions that in the embodiment of the present application, the encoding end can judge whether to allow the second transformation of the current block according to the first preset condition and the second preset condition. If the current block satisfies the first preset condition, and the current If the block also satisfies the second preset condition, the current block is allowed to undergo secondary transformation. Based on this, it may be determined whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block does not meet the first preset condition, it is prohibited to perform secondary transformation on the current block, so as to eliminate blocks that do not need to undergo secondary transformation, play a role in removing redundancy, and improve coding performance.

Embodiment 4: Referring to FIG. 3B, it is a schematic flow diagram of a decoding method, which can be applied to a decoding end. The method includes:

Step 321, acquire the coded bit stream of the current block.

Step 322, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the third preset condition, then determine whether the current block satisfies the fourth preset condition; if the current block satisfies the fourth preset condition, then The indication information is parsed from the coded bit stream, and whether to perform secondary inverse transformation on the current block is determined according to the indication information.

Exemplarily, if the value of the secondary transformation enabling flag (such as SecondaryTransformEnableFlag) is a preset value (such as 1), it is determined that the current block supports secondary transformation. When the value of the secondary transformation enable flag is a preset value, it means that the secondary transformation is allowed to be enabled at the sequence level or the frame level, that is, the current block is allowed to enable the secondary transformation. Therefore, the decoder determines that the current block supports the secondary transformation.

Exemplarily, if the value of the optional secondary transform enabling flag (such as AstEnableFlag) is a preset value (such as 1), it is determined that the current block supports AST technology. When the optional secondary transform enabling flag is a preset value, it means that the sequence level or frame level allows enabling the AST technology, that is, allowing the current block to enable the AST technology, therefore, the decoder determines that the current block supports the AST technology.

Exemplarily, if the current block does not support secondary transformation, the decoder determines not to perform secondary inverse transformation on the current block. If the current block supports secondary transformation, but the current block does not support AST technology, the decoder determines not to perform secondary inverse transformation on the current block.

If the current block supports secondary transformation and the current block supports AST technology, the decoding end judges whether the current block satisfies the third preset condition. If the current block does not satisfy the third preset condition, the decoding end determines not to perform the second inverse transform on the current block.

If the current block supports secondary transformation, and the current block supports AST technology, and the current block satisfies the third preset condition, then the decoder determines whether the current block satisfies the fourth preset condition. If the current block does not satisfy the fourth preset condition, it is determined whether to perform the second inverse transformation on the current block, without determining whether to perform the second inverse transformation on the current block according to the indication information in the coded bit stream.

If the current block supports secondary transformation, and the current block supports AST technology, and the current block satisfies the third preset condition, then the decoder determines whether the current block satisfies the fourth preset condition. If the current block satisfies the fourth preset condition, the decoding end parses the indication information from the encoded bit stream, and determines whether to perform secondary inverse transformation on the current block according to the indication information. For example, the encoder may explicitly encode indication information to indicate whether to perform secondary transformation on the current block. Based on this, the decoding end parses the indication information from the coded bit stream. If the indication information is the first indication information, determine to perform a second inverse transformation on the current block; the first indication information is used to indicate to perform a second transformation on the current block; or, if the indication information is the second indication information, determine No secondary inverse transformation is performed on the current block; the second indication information is used to indicate that no secondary transformation is performed on the current block.

In the above embodiment, the decoder needs to judge whether the current block meets the third preset condition. For the third preset condition, in a possible implementation manner, the third preset condition may include but not limited to the following conditions at least one:

Condition 1: the prediction mode of the current block is an intra prediction mode.

Condition 2: the current block is a luma block.

Condition 3: The initial transform coefficient matrix of the current block contains non-zero transform coefficients.

Exemplarily, if the initial transform coefficient matrix does not contain non-zero transform coefficients, the cbf flag of the current block is the first value (such as a value of 0), and if the initial transform coefficient matrix contains non-zero transform coefficients (that is, any one or multiple transform coefficients are not zero), then the cbf flag bit of the current block is the second value (such as value 1). Based on this, the decoder can parse out the cbf flag bit from the encoded bit stream. If the cbf flag bit is the first value, it is determined that the initial transform coefficient matrix does not contain non-zero transform coefficients, that is, all transform coefficients in the initial transform coefficient matrix are all zero. If the cbf flag bit is the second value, it is determined that the initial transform coefficient matrix contains non-zero transform coefficients, that is, all transform coefficients in the initial transform coefficient matrix are not zero.

Condition 4: The mode number of the intra prediction mode adopted by the current block is located in the first mode number interval, and the reference samples on the left outside the current block are available; or, the mode number of the intra prediction mode adopted by the current block is located in the second mode number interval , and the reference samples above and outside the current block are available. Exemplarily, the first mode number interval may include mode number 0 to mode number 2, mode number 13 to mode number 32, and mode number 44 to mode number 65. The second mode number interval may include mode number 0 to mode number 23, and mode number 34 to mode number 57. Of course, the above are only examples of the first mode number interval and the second mode number interval, and there is no limitation on this.

Condition 5: The current block uses the transform check (DCT2, DCT2) for initial inverse transform; or, the current block does not use the transform check (DST7, DST7) for initial inverse transform. Alternatively, istTuflag is equal to the first value (such as 0).

For example, if transform checks (DCT2, DCT2) are used for initial inverse transformation, condition 5 is satisfied, and if transform checks (DST7, DST7) are used for initial inverse transformation, condition 5 is not satisfied.

Exemplarily, the decoding end determines whether the current block uses the transformation check (DCT2, DCT2) to perform the initial inverse transformation, or uses the transformation check (DST7, DST7) to perform the initial inverse transformation, the following method can be adopted: the decoding end parses the encoded bit stream to obtain sr_x and sr_y, when sr_x is greater than or equal to 16 or sr_y is greater than or equal to 16, the decoder can directly determine that the current block uses the transform check (DCT2, DCT2) for initial inverse transform. When sr_x is less than 16 and sr_y is less than 16, the decoding end can determine the parity of the number of all non-zero transform coefficients in the target transform coefficient matrix. If the parity is odd, the decoder can determine that the current block uses transform check (DST7, DST7) for initial inverse transform; if the parity is even, the decoder can determine that the current block uses transform check (DCT2, DCT2) initial inverse transformation.

Of course, the above method is just an example, without limitation, as long as it can be determined whether the current block uses the transform check (DCT2, DCT2) for the initial inverse transform or uses the transform check (DST7, DST7) for the initial inverse transform.

Exemplarily, the third preset condition may only include condition 1 and condition 2. Alternatively, the third preset condition may only include condition 1, condition 2 and condition 5. Alternatively, the third preset condition may include condition 1, condition 2, and condition 5. On this basis, the third preset condition may also include condition 4 and/or condition 5. Of course, the above is just an example and is not limited to this .

Exemplarily, the third preset condition is similar to the first preset condition, which will not be repeated here.

In the above embodiments, it is necessary to determine whether the current block satisfies the fourth preset condition. For the fourth preset condition, in a possible implementation manner, the fourth preset condition may include but not limited to at least one of the following conditions :

Condition a: the current block uses the intra prediction filtering mode, or the current block does not use the intra prediction filtering mode.

Condition b: the current block supports the DT division mode; or, the current block does not support the DT division mode.

Condition c: the width and height of the current block satisfy the first size condition. For example, if the width of the current block is less than or equal to the first value, and the height of the current block is less than or equal to the second value, then the width and height of the current block satisfy the first size condition.

For example, the first numerical value can be 32 or 64, and the second numerical value can be 32 or 64. Of course, 32 and 64 are just examples, which are not limited. For example, the first numerical value can also be 128, and the second numerical value can also be 128. Wait. For example, the first numerical value is 32 and the second numerical value is 32, or the first numerical value may be 64 and the second numerical value may be 64.

Exemplarily, the fourth preset condition may only include condition a. Alternatively, the fourth preset condition may only include condition a and condition b. Alternatively, the fourth preset condition may only include condition a and condition c. Alternatively, the fourth preset condition may only include condition b. Alternatively, the fourth preset condition may only include condition b and condition c. Alternatively, the fourth preset condition may include condition a, condition b and condition c. Of course, the above is just an example, without limitation.

Exemplarily, the fourth preset condition is similar to the second preset condition, which will not be repeated here.

Step 323, parse out the target transform coefficient matrix of the current block from the coded bit stream.

Step 324: If the current block is subjected to the second inverse transformation, perform the second inverse transformation on the transformation coefficients in the specified area in the upper left corner of the target transformation coefficient matrix to obtain the initial transformation coefficient matrix. If no secondary inverse transform is performed on the current block, the target transform coefficient matrix is determined as the initial transform coefficient matrix. So far, the initial transformation coefficient matrix of the current block is obtained.

Exemplarily, performing a second inverse transform on the transform coefficients of the specified area in the upper left corner of the target transform coefficient matrix may include but not limited to: if the width and height of the current block meet the second size condition, then based on the first specified The size of the secondary inverse transformation matrix (the secondary inverse transformation matrix and the secondary transformation matrix used in the secondary transformation of the encoding end can be the same), the transformation of the first specified area in the upper left corner of the target transformation coefficient matrix Coefficients undergo a second inverse transformation. If the width and height of the current block do not meet the second size condition, the second inverse transform can be performed on the transform coefficients of the second specified area located in the upper left corner of the target transform coefficient matrix based on the second specified size of the second inverse transform matrix .

Exemplarily, the size of the first specified area is the first specified size, the size of the second specified area is the second specified size, and the first specified size is larger than the second specified size. For example, the first specified size is 8*8; the second specified size is 4*4. Of course, the above are only examples of the first specified size and the second specified size, and there is no limitation to the first specified size and the second specified size.

The width and height of the current block meet the second size condition, which may include but not limited to: the width of the current block is greater than or equal to the third value, the height of the current block is greater than or equal to the fourth value; or, the width of the current block is greater than or equal to the fifth value, The height of the current block is greater than the sixth value; or, the width of the current block is greater than the seventh value, and the height of the current block is greater than or equal to the eighth value.

Exemplarily, the above-mentioned third value, fourth value, fifth value, sixth value, seventh value, and eighth value can all be configured according to experience, and these values are not limited. For example, the third numerical value may be 8 or 16, the fourth numerical value may be 8 or 16, the fifth numerical value may be 8 or 16, and the sixth numerical value may be 8 or 16. The seventh numerical value may be 8 or 16, and the eighth numerical value may be 8 or 16.

In summary, if the width and height of the current block satisfy the second size condition (such as the width of the current block is greater than or equal to 8, and the height of the current block is greater than or equal to 8), then based on the secondary inverse transformation matrix with a size of 8*8, A second inverse transform is performed on the transform coefficients of the first designated area (ie, the area with a size of 8*8) located at the upper left corner in the target transform coefficient matrix. If the width and height of the current block do not satisfy the second size condition, then based on the quadratic inverse transform matrix with a size of 4*4, the second specified area located in the upper left corner in the target transform coefficient matrix (that is, the area with a size of 4*4 Area) transform coefficients are subjected to a second inverse transform.

Step 325: Perform an initial inverse transformation on the initial transformation coefficient matrix to obtain a residual coefficient matrix corresponding to the current block.

Exemplarily, referring to condition 8 in the above-mentioned embodiment, the decoder can determine whether the current block uses the transform check (DCT2, DCT2) to perform the initial inverse transform, or uses the transform check (DST7, DST7) to perform the initial inverse transform. The method will not be repeated. If transform check (DCT2, DCT2) is used to perform initial inverse transform, the decoding end may use transform check (DCT2, DCT2) to perform initial inverse transform on the initial transform coefficient matrix to obtain the residual coefficient matrix corresponding to the current block. Alternatively, if the transformation check (DST7, DST7) is used to perform the initial inverse transformation, the decoding end may use the transformation check (DST7, DST7) to perform the initial inverse transformation on the initial transformation coefficient matrix to obtain the residual coefficient matrix corresponding to the current block.

Step 326, determine the reconstruction value of the current block according to the residual coefficient matrix.

For example, the reference block corresponding to the current block may be determined, and for each pixel of the current block, the reference point corresponding to the pixel is determined from the reference block, and the residual coefficient corresponding to the pixel is determined from the residual coefficient matrix value. According to the pixel value of the reference point and the residual coefficient value, the reconstructed value of the pixel point can be obtained. The reconstruction values corresponding to all the pixels of the current block can form the reconstruction value of the current block. Of course, the above manner is only an example, and is not limited thereto.

To sum up, the decoder can obtain istTuflag according to the parity of the number of all non-zero transform coefficients in the target transform coefficient matrix, and sr_x and sr_y. If istTuflag is equal to 0, it means that the current block uses the transform check (DCT2, DCT2) for initial inverse transform; if istTuflag is equal to 1, it means that the current block uses the transform check (DST7, DST7) for initial inverse transform. When istTuflag is equal to 0 and needs to perform a second inverse transformation on the current block, the decoder performs a second inverse transformation on the target transformation coefficient matrix to obtain the initial transformation coefficient matrix, and then performs an initial inverse transformation on the initial transformation coefficient matrix to obtain the residual Difference coefficient matrix. When istTuflag is equal to 0 and there is no need to perform a second inverse transform on the current block, the decoder determines the target transform coefficient matrix as the initial transform coefficient matrix, performs initial inverse transform on the initial transform coefficient matrix, and obtains the residual coefficient matrix. When istTuflag is equal to 1, the decoding end determines the target transformation coefficient matrix as the initial transformation coefficient matrix, performs initial inverse transformation on the initial transformation coefficient matrix, and obtains the residual coefficient matrix.

Then, the reconstruction value of the current block can be determined by the residual coefficient matrix, and the reconstruction process will not be repeated here.

It can be seen from the above technical solutions that in the embodiment of the present application, the decoding end can judge whether to allow the second inverse transformation of the current block according to the third preset condition and the fourth preset condition, for example, if the current block satisfies the third preset condition condition, and the current block also satisfies the fourth preset condition, then the current block is allowed to be subjected to secondary inverse transformation, and whether to perform secondary inverse transformation on the current block is determined according to the coded bit stream of the current block. If the current block does not meet the third preset condition, the second inverse transformation of the current block is prohibited, that is, the decoding end directly determines that the current block does not undergo the second inverse transformation, thereby eliminating the blocks that do not need to undergo the second transformation. The role of de-redundancy, thereby improving the coding performance.

Embodiment 5: Referring to FIG. 4A, it is a schematic flow diagram of an encoding method, which can be applied to the encoding end. The method includes:

Step 411, acquire the residual coefficient matrix corresponding to the current block.

Step 412: Perform an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix.

Step 413, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the first preset condition, then determine whether the current block satisfies the second preset condition; if the current block satisfies the second preset condition, then According to the rate-distortion cost value corresponding to the current block, it is determined whether to perform secondary transformation on the current block.

Step 414: If the current block is to be transformed twice, then the transform coefficients in the specified area in the upper left corner of the initial transform coefficient matrix are transformed twice to obtain the target transform coefficient matrix; if the current block is not transformed twice, then The initial transform coefficient matrix is determined as the target transform coefficient matrix. So far, the target transformation coefficient matrix of the current block can be obtained.

For example, step 411-step 414 may refer to step 311-step 314, which will not be repeated here.

Step 415: Determine the target feature of the target transform coefficient matrix according to the transform coefficients in the target transform coefficient matrix.

Exemplarily, the target feature may be target parity, and the target parity may include but not limited to: the parity of the number of all odd transform coefficients in the target transform coefficient matrix; for example, if there are 30 odd transform coefficients in the target transform coefficient matrix transform coefficients, the parity of all odd transform coefficients is an even number, that is, the target parity is an even number. Or, the parity of the number of all even transform coefficients in the target transform coefficient matrix; for example, if there are 25 even transform coefficients in the target transform coefficient matrix, the parity of the number of all even transform coefficients is odd, that is, the target Parity is odd. Or, the parity of the sum of the absolute values of all the transform coefficients in the target transform coefficient matrix; for example, if the sum of the absolute values of all the transform coefficients in the target transform coefficient matrix is odd, then the target parity is Odd, if the sum result is even, then the target parity is even. Of course, the above target parity is just a few examples, which is not limited. For example, the target parity can also be the parity of the number of coefficients with an absolute value of 1, the parity of the last non-zero transformation coefficient, and so on.

The target feature can also be the sign of the target, and the sign of the target can include but not limited to: the sign of the last non-zero transform coefficient in the target transform coefficient matrix; or, the first non-zero transform coefficient in the target transform coefficient matrix The positive or negative of the transform coefficient; or, the positive or negative of the Nth non-zero transform coefficient in the target transform coefficient matrix, the value of N can be arbitrarily configured according to actual needs, and there is no limitation on this. Certainly, the above-mentioned target sign or negative are just a few examples, and there is no limitation on this target sign or negative, for example, the target sign or negative may be the sign or negative of the sum of the absolute values of all the transform coefficients in the target transform coefficient matrix.

Of course, target parity and target positive/negative are just two examples of target features, and there is no limitation on the target features. For the convenience of description, target parity is used as an example for description in the following, and other target features are processed in a similar manner.

Step 416: Determine whether to adjust the target transformation coefficient matrix according to the target feature and the target value of the preset flag bit of the current block. Exemplarily, when the target value of the preset flag bit is the first value, it is used to indicate that the current block is to be transformed twice, and when the target value of the preset flag bit is the second value, it is used to indicate that the current block is transformed Blocks are not retransformed.

Exemplarily, after the encoding end determines whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block, if the current block is subjected to secondary transformation, the target value of the preset flag can be set to the first A value to indicate that the current block should be retransformed. If the second transformation is not performed on the current block, the target value of the preset flag bit may be set as the second value to indicate that the second transformation is not performed on the current block. For example, the first value may be 1, and the second value may be 0; or, the first value may be 0, and the second value may be 1; of course, the above is only an example, and is not limited thereto.

In a possible implementation manner, the encoding end and the decoding end can agree on the relationship between the target parity and the secondary transformation. For example, when the target parity is odd, it means performing secondary transformation on the current block, and when the target parity is even , indicating that the current block is not subjected to secondary transformation, that is, the relationship between "odd number" and secondary transformation, "even number" and no secondary transformation is agreed.

In the above case, if the target parity is an odd number, and the target value of the preset flag bit of the current block is the first value, it is determined not to adjust the target transformation coefficient matrix. For example, if the target parity is An odd number means that the current block is to be transformed twice, and the target value of the preset flag bit is the first value, which also means that the current block is transformed twice, so the target transformation coefficient matrix is not adjusted. Or, if the target parity is an odd number, and the target value of the preset flag bit of the current block is the second value, it is determined to adjust the target transformation coefficient matrix. For example, if the target parity is an odd number, it means The second transformation is performed on the current block, and the target value of the preset flag bit is the second value, which means that the second transformation is not performed on the current block. Therefore, the target transformation coefficient matrix is adjusted. Or, if the target parity is an even number, and the target value of the preset flag bit of the current block is the first value, it is determined to adjust the target transform coefficient matrix. For example, if the target parity is an even number, it means The current block is not subjected to secondary transformation, and the target value of the preset flag bit is the first value, indicating that the current block is subjected to secondary transformation. Therefore, the target transformation coefficient matrix is adjusted. Or, if the target parity is an even number, and the target value of the preset flag bit of the current block is the second value, it is determined not to adjust the target transform coefficient matrix. For example, if the target parity is an even number, then Indicates that no secondary transformation is performed on the current block, and the target value of the preset flag bit is the second value, which indicates that secondary transformation is not performed on the current block, therefore, the target transformation coefficient matrix is not adjusted.

In another possible implementation, the encoder and the decoder can agree on the relationship between the target parity and the secondary transformation. For example, when the target parity is an even number, it means that the current block is subjected to secondary transformation, and the target parity is an odd number. When , it means that the current block will not be transformed twice, that is, the relationship between "even number" and second transformation, and "odd number" and no second transformation will be agreed.

In the above case, if the target parity is an even number, and the target value of the preset flag bit of the current block is the first value, it is determined not to adjust the target transform coefficient matrix. For example, if the target parity is An even number means that the current block is to be transformed twice, and the target value of the preset flag bit is the first value, which also means that the current block is transformed twice, so the target transformation coefficient matrix is not adjusted. Or, if the target parity is an even number, and the target value of the preset flag bit of the current block is the second value, it is determined to adjust the target transform coefficient matrix. For example, if the target parity is an even number, it means The second transformation is performed on the current block, and the target value of the preset flag bit is the second value, which means that the second transformation is not performed on the current block. Therefore, the target transformation coefficient matrix is adjusted. Or, if the target parity is an odd number, and the target value of the preset flag bit of the current block is the first value, it is determined to adjust the target transformation coefficient matrix. For example, if the target parity is an odd number, it means The current block is not subjected to secondary transformation, and the target value of the preset flag bit is the first value, indicating that the current block is subjected to secondary transformation. Therefore, the target transformation coefficient matrix is adjusted. Or, if the target parity is an odd number, and the target value of the preset flag bit of the current block is the second value, it is determined not to adjust the target transform coefficient matrix. For example, if the target parity is an odd number, then Indicates that no secondary transformation is performed on the current block, and the target value of the preset flag bit is the second value, which indicates that secondary transformation is not performed on the current block, therefore, the target transformation coefficient matrix is not adjusted.

Exemplarily, if the target transformation coefficient matrix is not adjusted, step 417 is performed.

Exemplarily, if the target transformation coefficient matrix is adjusted, step 418 is performed.

Step 417: Encode the target transformation coefficient matrix to obtain an encoded bit stream of the current block.

Step 418: Adjust the transform coefficients in the target transform coefficient matrix, so that the adjusted target feature of the target transform coefficient matrix matches the target value of the preset flag bit. After the transformation coefficients in the target transformation coefficient matrix are adjusted, the adjusted target transformation coefficient matrix is encoded to obtain an encoded bit stream of the current block.

For example, if the encoding end and the decoding end agree on the relationship between "odd number" and secondary transformation, "even number" and no secondary transformation, and the target parity is odd, the target value of the preset flag bit of the current block is the first Two values, based on this, an odd transformation coefficient in the target transformation coefficient matrix can be adjusted so that the odd transformation coefficient is changed to an even number, such as adding 1 to the odd transformation coefficient, or subtracting 1, after the change The transformation coefficient of is an even number, not 0. Or, adjust an even transformation coefficient in the target transformation coefficient matrix so that the even transformation coefficient is changed to an odd number, such as adding 1 to or subtracting 1 from the even transformation coefficient, and the changed transformation coefficient is an odd number, not 0.

After the above processing, the target parity of the adjusted target transformation coefficient matrix is an even number, "even number" means that the current block will not be transformed twice, and the target value of the preset flag bit is the second value, which means that the current block will not be transformed twice. A secondary transformation is performed, that is, the adjusted target parity of the target transform coefficient matrix matches the target value of the preset flag bit.

For another example, if the target parity is even, the target value of the preset flag bit of the current block is the first value. Based on this, an odd transform coefficient in the target transform coefficient matrix can be adjusted so that the odd transform The coefficient is changed to an even number, such as adding 1 or subtracting 1 to the odd transform coefficient. Alternatively, an even transformation coefficient in the target transformation coefficient matrix is adjusted so that the even transformation coefficient is changed to an odd number, such as adding 1 to or subtracting 1 from the even transformation coefficient. After the above processing, the target parity of the adjusted target transformation coefficient matrix is an odd number, "odd number" indicates that the current block is subjected to secondary transformation, and the target value of the preset flag bit is the first value, indicating that the current block is subjected to secondary transformation. transformation, that is, the adjusted target parity of the target transform coefficient matrix matches the target value of the preset flag bit.

For another example, if the encoding end and the decoding end agree on the relationship between "even number" and secondary transformation, and "odd number" and no secondary transformation, the adjustment process in this case is similar to the above adjustment method, and will not be repeated here. .

To sum up, the encoder can implicitly indicate whether to perform secondary transformation on the current block through the target characteristics (such as parity) of the target transformation coefficient matrix, instead of explicitly indicating whether to perform secondary transformation on the current block through indication information. transform.

Exemplarily, in the encoding process, if the current block uses the (DST7, DST7) transform kernel for initial transformation, the target transformation coefficient matrix after transformation and quantization needs to satisfy: sr_x and sr_y are less than 16, and all in the target transformation coefficient matrix The number of non-zero transform coefficients is an odd number, which is used to indicate that the current block uses the (DST7, DST7) transform kernel for initial transform. Or, if the current block uses the (DCT2, DCT2) transform kernel for initial transformation, the target transform coefficient matrix after transformation and quantization needs to satisfy: sr_x or sr_y is greater than or equal to 16, or, all non-zero transform coefficients in the target transform coefficient matrix The number is an even number, which is used to indicate that the current block uses the (DCT2, DCT2) transform kernel for initial transformation.

Exemplarily, if the current block supports secondary transformation, the current block supports AST technology, the current block satisfies the first preset condition, and the current block satisfies the second preset condition, it is determined according to the rate-distortion cost value corresponding to the current block After the secondary transformation is performed, step 415 to step 418 are performed to implicitly indicate that the current block undergoes secondary transformation or does not perform secondary transformation.

If the current block does not support secondary transformation, it is determined that secondary transformation is not performed on the current block, but steps 415 to 418 do not need to be performed, that is, there is no need to implicitly indicate that the current block does not undergo secondary transformation. Or, if the current block supports secondary transformation, but the current block does not support AST technology, it is determined that the current block is not subjected to secondary transformation, but steps 415-418 do not need to be performed, that is, it is not necessary to implicitly indicate that the current block does not perform secondary transformation. Or, if the current block supports secondary transformation, the current block supports AST technology, and the current block does not meet the first preset condition, it is determined that the current block is not subjected to secondary transformation, but steps 415-418 do not need to be performed, that is, no In an implicit way, it indicates that the current block does not undergo secondary transformation. Or, if the current block supports secondary transformation, the current block supports AST technology, the current block satisfies the first preset condition, and the current block does not meet the second preset condition, then it is determined to perform secondary transformation on the current block, but no steps need to be performed 415-step 418, that is, it is not necessary to implicitly represent the current block to perform secondary transformation.

It can be seen from the above technical solutions that in the embodiment of the present application, the encoding end can judge whether to allow the second transformation of the current block according to the first preset condition and the second preset condition. If the current block satisfies the first preset condition, and the current If the block also satisfies the second preset condition, the current block is allowed to undergo secondary transformation. Based on this, it may be determined whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block does not meet the first preset condition, it is prohibited to perform secondary transformation on the current block, so as to eliminate blocks that do not need to undergo secondary transformation, play a role in removing redundancy, and improve coding performance.

Embodiment 6: Referring to FIG. 4B, it is a schematic flow diagram of a decoding method, which can be applied to a decoding end. The method includes:

Step 421, acquire the coded bit stream of the current block, and parse the target transformation coefficient matrix of the current block from the coded bit stream.

Step 422, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the third preset condition, then determine whether the current block satisfies the fourth preset condition; if the current block satisfies the fourth preset condition, then Target features of the target transform coefficient matrix are determined based on the transform coefficients in the target transform coefficient matrix.

Exemplarily, the target feature may include but not limited to target parity or target sign-off. Of course, target parity and target positive/negative are just two examples of target features, and there is no limitation on the target features. Wherein, the target parity may include but not limited to: the parity of the number of all odd transform coefficients in the target transform coefficient matrix; or, the parity of the number of all even transform coefficients in the target transform coefficient matrix; or, the target Parity of the sum of absolute values of all transform coefficients in the transform coefficient matrix. Of course, the above are just a few examples, which are not limited. For example, the target parity may also be the parity of the number of coefficients with an absolute value of 1, the parity of the last non-zero transform coefficient, and the like. The target sign or negativity may include, but is not limited to: the sign of the last non-zero transform coefficient in the target transform coefficient matrix; or, the sign of the first non-zero transform coefficient in the target transform coefficient matrix; or, the target The positive or negative of the Nth non-zero transform coefficient in the transform coefficient matrix, and the value of N can be configured arbitrarily. Of course, the above-mentioned target signabilities are just a few examples, and there is no limitation to this. For example, the target signabilities may be the signabilities of the sum of the absolute values of all transform coefficients in the target transform coefficient matrix.

For the convenience of description, target parity is used as an example for description in the following, and other target features are processed in a similar manner.

After the target transform coefficient matrix is obtained, the target parity of the target transform coefficient matrix can be determined according to the transform coefficients in the target transform coefficient matrix. For example, assuming that the target parity is the parity of the number of all odd transform coefficients in the target transform coefficient matrix, if there are 30 odd transform coefficients in the target transform coefficient matrix, the target parity is an even number.

Step 423: Determine the target value of the preset flag bit of the current block according to the target feature.

Exemplarily, when the target value of the preset flag bit is the first value, it is used to indicate that a second inverse transformation is performed on the current block, and when the target value of the preset flag bit is the second value, it is used to Indicates that no secondary inverse transform is to be performed on the current block.

In a possible implementation manner, the decoding end and the encoding end can agree on the relationship between the target parity and the second inverse transformation. For example, when the target parity is an odd number, it means that the current block is subjected to the second inverse transformation. The target parity is When the number is even, it means that the current block will not be subjected to the second inverse transformation, that is, the relationship between "odd number" and the second inverse transformation, "even number" and no second inverse transformation. If the target parity of the target transformation coefficient matrix is an odd number, then determine that the target value of the preset flag bit of the current block is the first value. Exemplarily, if the target parity is an odd number, it means performing a second inverse transformation on the current block, and the target value of the preset flag bit is the first value, which also means performing a second inverse transformation on the current block, that is, the target parity The property matches the target value. If the target parity of the target transformation coefficient matrix is an even number, then determine the target value of the preset flag bit of the current block as the second value. Exemplarily, if the target parity is an even number, it means that the second inverse transformation is not performed on the current block, and the target value of the preset flag bit is the second value, which also means that the second inverse transformation is not performed on the current block, namely The target parity matches the target value.

In another possible implementation, the decoding end and the encoding end can agree on the relationship between the target parity and the second inverse transformation. For example, when the target parity is even, it means that the current block is subjected to the second inverse transformation, and the target parity When it is an odd number, it means that the current block will not be subjected to secondary inverse transformation, that is, the relationship between "even number" and secondary inverse transformation, and "odd number" and no secondary inverse transformation. If the target parity of the target transformation coefficient matrix is an even number, then determine that the target value of the preset flag bit of the current block is the first value. Exemplarily, if the target parity is an even number, it means performing a second inverse transform on the current block, and the target value of the preset flag bit is the first value, which also means performing a second inverse transform on the current block, that is, the target parity The property matches the target value. If the target parity of the target transformation coefficient matrix is an odd number, then determine the target value of the preset flag bit of the current block as the second value. Exemplarily, if the target parity is an odd number, it means that the second inverse transformation is not performed on the current block, and the target value of the preset flag bit is the second value, which also means that the second inverse transformation is not performed on the current block, namely The target parity matches the target value.

Step 424: Determine whether to perform secondary inverse transformation on the current block according to the target value of the preset flag bit.

Exemplarily, if the target value of the preset flag bit is the first value, it can be determined to perform a second inverse transform on the current block; if the target value of the preset flag bit is the second value, it can be determined to The current block does not undergo secondary inverse transformation.

Step 425: If the current block is subjected to the second inverse transformation, perform the second inverse transformation on the transformation coefficients in the specified area in the upper left corner of the target transformation coefficient matrix to obtain the initial transformation coefficient matrix. If no secondary inverse transform is performed on the current block, the target transform coefficient matrix is determined as the initial transform coefficient matrix. So far, the initial transformation coefficient matrix of the current block is obtained.

Step 426: Perform an initial inverse transformation on the initial transformation coefficient matrix to obtain a residual coefficient matrix corresponding to the current block.

Step 427, determine the reconstruction value of the current block according to the residual coefficient matrix.

Exemplarily, step 421-step 422, step 425-step 427, refer to Embodiment 4, and details are not repeated here.

To sum up, the decoder can obtain istTuflag according to the parity of the number of all non-zero transform coefficients in the target transform coefficient matrix, and sr_x and sr_y. If istTuflag is equal to 0, it means that the current block uses the transform check (DCT2, DCT2) for initial inverse transform; if istTuflag is equal to 1, it means that the current block uses the transform check (DST7, DST7) for initial inverse transform. When istTuflag is equal to 0 and the current block satisfies the third preset condition and the fourth preset condition, the decoder determines whether to perform a second inverse transform on the target transform coefficient matrix according to the target parity of the target transform coefficient matrix. When istTuflag is equal to 0 and the current block does not meet the third preset condition, the decoder determines not to perform a second inverse transform on the target transform coefficient matrix. When istTuflag is equal to 0, and the current block satisfies the third preset condition but does not satisfy the fourth preset condition, the decoder determines to perform a second inverse transform on the target transform coefficient matrix. When istTuflag is equal to 1, the decoder determines that the current block does not meet the third preset condition, and does not perform a second inverse transform on the target transform coefficient matrix.

If it is necessary to perform a second inverse transformation on the target transformation coefficient matrix, the decoder performs a second inverse transformation on the target transformation coefficient matrix to obtain the initial transformation coefficient matrix, and then performs an initial inverse transformation on the initial transformation coefficient matrix to obtain the residual coefficient matrix . If the target transformation coefficient matrix is not subjected to secondary inverse transformation, the decoding end determines the target transformation coefficient matrix as the initial transformation coefficient matrix, performs initial inverse transformation on the initial transformation coefficient matrix, and obtains the residual coefficient matrix.

Then, the reconstruction value of the current block can be determined by the residual coefficient matrix, and the reconstruction process will not be repeated here.

It can be seen from the above technical solutions that in the embodiment of the present application, the decoding end can judge whether to allow the second inverse transformation of the current block according to the third preset condition and the fourth preset condition, for example, if the current block satisfies the third preset condition condition, and the current block also satisfies the fourth preset condition, then the current block is allowed to be subjected to secondary inverse transformation, and whether to perform secondary inverse transformation on the current block is determined according to the coded bit stream of the current block. If the current block does not meet the third preset condition, the second inverse transformation of the current block is prohibited, that is, the decoding end directly determines that the current block does not undergo the second inverse transformation, thereby eliminating the blocks that do not need to undergo the second transformation. The role of de-redundancy, thereby improving the coding performance.

Embodiment 7: In Embodiment 1-Embodiment 4, for the encoding end, RDO can be used to determine whether to perform secondary transformation on the current block, and explicitly encode indication information (such as secTuflag) to indicate whether to perform the current block Perform secondary transformation. For example, when the value of secTuflag is the first value, it indicates that the current block is subjected to secondary transformation; when the value of secTuflag is the second value, it indicates that the current block is not subjected to secondary transformation. For the decoding end, after obtaining the coded bit stream, the indication information (such as secTuflag) of the current block can be parsed from the coded bit stream, and if the value of secTuflag is the first value, it is determined to perform secondary secondary inverse transform, if the value of secTuflag is the second value, it is determined not to perform secondary inverse transform on the current block. The implementation manner of the encoding instruction information will be described below in conjunction with several specific application scenarios.

Application Scenario 1: For the encoding side, if the current block supports secondary transformation, the current block supports AST technology, and the current block meets the first preset condition, then it is judged whether the current block meets the second preset condition. two preset conditions, it may be determined whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block does not satisfy the second preset condition, it may be determined to perform secondary transformation on the current block. Then, when determining whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block, if the current block is subjected to secondary transformation, secTuflag (that is, indication information) may be added to the coded bitstream of the current block , and the value of secTuflag is the first value, and the first value is used to indicate that a second transformation is performed on the current block. If the current block is not to be transformed twice, secTuflag is added to the coded bit stream of the current block, and the value of secTuflag is the second value, and the second value is used to indicate that the current block is not to be transformed twice.

Exemplarily, if the current block uses the transformation kernel (DCT2, DCT2) for initial transformation, the target transformation coefficient matrix after transformation and quantization needs to meet: sr_x is greater than or equal to 16 or sr_y is greater than or equal to 16, or, all of the target transformation coefficient matrix The number of non-zero transform coefficients is an even number, which is used to indicate that the current block uses (DCT2, DCT2) transform kernel for initial transform. Or, if the current block uses the (DST7, DST7) transform kernel for initial transformation, the target transform coefficient matrix after transformation and quantization needs to satisfy: sr_x and sr_y are less than 16, and the number of all non-zero transform coefficients in the target transform coefficient matrix It is an odd number, used to indicate that the current block uses (DST7, DST7) transform kernel for initial transformation.

For the decoder, istTuflag is obtained according to the parity of the number of all non-zero transform coefficients in the target transform coefficient matrix, and sr_x and sr_y. If istTuflag is equal to 0, it means that the current block uses the transform check (DCT2, DCT2) for initial inverse transform; if istTuflag is equal to 1, it means that the current block uses the transform check (DST7, DST7) for initial inverse transform.

If the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the third preset condition, then it is judged whether the current block satisfies the fourth preset condition, and if the current block does not meet the fourth preset condition, the decoder determines The second inverse transform is performed on the current block, and the decoding indication information (secTuflag) is not required at this time. If the current block satisfies the fourth preset condition, the decoder can parse the secTuflag (ie, indication information) from the encoded bit stream, and determine whether to perform a second inverse transform on the current block based on the secTuflag. For example, if the value of secTuflag is the first If it takes a value, it is determined to perform a second inverse transformation on the current block. If the value of secTuflag is the second value, it is determined not to perform the second inverse transform on the current block.

If a second inverse transform is performed on the current block, a second inverse transform is performed on the target transform coefficient matrix to obtain an initial transform coefficient matrix. If no secondary inverse transform is performed on the current block, the target transform coefficient matrix is determined as the initial transform coefficient matrix. Then, the residual block can be obtained according to the initial transformation coefficient matrix, and the reconstruction block can be obtained by adding the residual block and the prediction block.

In the above embodiment, the first preset condition or the third preset condition may include but not limited to: the prediction mode of the current block is intra-frame prediction mode; the current block is a luma block; the initial transformation coefficient matrix of the current block contains non- Zero transform coefficient (for example, the cbf flag of the current block is the second value 1). The second preset condition or the fourth preset condition may include but not limited to: the current block does not use the intra prediction filtering mode (ipf_flag=0); the current block does not support the DT division mode (or the current block supports the DT division mode). Optionally, the second preset condition or the fourth preset condition may further include: the width and height of the front block satisfy the first size condition. For example, the width W and height H of the current block satisfy: W<=n1 and H<=n2. Exemplarily, n1 and n2 may be: n1=n2=32, or, n1=n2=64.

Application scenario 2: For the encoding side, if the current block supports secondary transformation, the current block supports AST technology, and the current block meets the first preset condition, it needs to judge whether the current block meets the second preset condition. If the current block meets the In the second preset condition, it may be determined whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. Then, if the current block is re-transformed, secTuflag (that is, indication information) can be added to the coded bitstream of the current block, and the value of secTuflag is the first value, and the first value is used to indicate that the current block Perform a secondary transformation. If the current block is not to be transformed twice, secTuflag is added to the coded bit stream of the current block, and the value of secTuflag is the second value, and the second value is used to indicate that the current block is not to be transformed twice.

For the decoding end, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the third preset condition, then it is judged whether the current block satisfies the fourth preset condition, and if the current block satisfies the fourth preset condition , then parse out secTuflag (that is, the indication information) from the encoded bit stream, and determine whether to perform a second inverse transform on the current block based on secTuflag, for example, if the value of secTuflag is the first value, then determine to perform a second inverse transform on the current block inverse transformation. If the value of secTuflag is the second value, it is determined not to perform the second inverse transform on the current block.

In the above embodiment, the first preset condition or the third preset condition may include but not limited to: the prediction mode of the current block is intra-frame prediction mode; the current block is a luma block; the initial transformation coefficient matrix of the current block contains non- Zero transform coefficient (for example, the cbf flag of the current block is the second value 1). The second preset condition or the fourth preset condition may include but not limited to: the current block uses the intra prediction filtering mode (ipf_flag=1); the current block does not support the DT division mode (or the current block supports the DT division mode). Optionally, the second preset condition or the fourth preset condition may further include: the width and height of the front block satisfy the first size condition. For example, the width W and height H of the current block satisfy: W<=n1 and H<=n2. Exemplarily, n1 and n2 may be: n1=n2=32, or, n1=n2=64.

Application Scenario 3: On the basis of Application Scenario 1 or Application Scenario 2, for the encoding side, if it is determined that the current block needs to be transformed twice, then the current block is transformed twice based on the size constraints of the current block. For example, if the width and height of the current block satisfy the second size condition, a second transform is performed on the transform coefficients of the first specified area (such as 8*8 area) located in the upper left corner in the initial transform coefficient matrix. If the width and height of the current block do not satisfy the second size condition, perform secondary transformation on the transformation coefficients of the second specified area (eg 4*4 area) located in the upper left corner in the initial transformation coefficient matrix.

Exemplarily, when the encoding end performs secondary transformation on the transformation coefficients of the first specified area (such as the 8*8 area) located in the upper left corner, the secondary transformation matrix used is an 8*8 secondary transformation matrix. When performing secondary transformation on the transformation coefficients of the second designated area (eg, 4*4 area) located in the upper left corner, the secondary transformation matrix used is a 4*4 secondary transformation matrix.

For the decoding end, if it is determined that the current block needs to be subjected to the second inverse transformation, then the current block may be subjected to the second inverse transformation based on the size restriction condition of the current block. For example, if the width and height of the current block satisfy the second size condition, a second inverse transform may be performed on the transform coefficients of the first specified area (eg, 8*8 area) located in the upper left corner of the target transform coefficient matrix. If the width and height of the current block do not satisfy the second size condition, a second inverse transform may be performed on the transform coefficients of the second specified area (eg, 4*4 area) located in the upper left corner of the target transform coefficient matrix. Exemplarily, when the decoding end performs a second inverse transform on the transform coefficients of the first designated area (such as an 8*8 area) located in the upper left corner, the second inverse transform matrix used is an 8*8 second inverse transform matrix . When the second inverse transformation is performed on the transformation coefficients of the second specified area (such as 4*4 area) located in the upper left corner, the second inverse transformation matrix used is a 4*4 second inverse transformation matrix.

Exemplarily, the above-mentioned 8*8 secondary transformation matrix and the 8*8 secondary inverse transformation matrix may be the same, that is, they are the same matrix. The above-mentioned 4*4 secondary transformation matrix and the 4*4 secondary inverse transformation matrix may be the same.

Exemplarily, the width W and height H of the current block satisfy the second size condition, which may be at least one of the following conditions:

Condition 1: W>=n3, and H>=n4, for example, W>=8, H>=8.

Condition 2: W>=n5, and H>n6, for example, W>=8, and H>8.

Condition 3: W>n7, and H>=n8, for example, W>8, and H>=8.

Embodiment 8: In Embodiment 1, Embodiment 2, Embodiment 5, and Embodiment 6, for the encoding end, it can be determined whether to perform secondary transformation on the current block through RDO, and through the target transformation coefficient matrix Features (such as target parity, etc.) to implicitly indicate whether to perform secondary transformation on the current block, instead of explicitly indicating whether to perform secondary transformation on the current block through indication information. For the decoding end, after obtaining the coded bit stream, the target transformation coefficient matrix is parsed from the coded bit stream, and the target value of the preset flag bit of the current block is determined according to the target characteristics of the target transformation coefficient matrix (such as secTuflag) , that is, deduce the target value of the preset flag bit of the current block. If the target value of the preset flag bit is the first value, it is determined to perform a second inverse transform on the current block. If the target value of the preset flag bit is the second value, it is determined not to perform the second inverse transform on the current block. The implementation of the implicit representation is described below in conjunction with several specific application scenarios.

Application scenario 4: For the encoding side, if the current block supports secondary transformation, the current block supports AST technology, and the current block meets the first preset condition, it needs to judge whether the current block meets the second preset condition. The second preset condition is to determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block does not satisfy the second preset condition, it is determined to perform secondary transformation on the current block. Then, when determining whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block, if the current block is subjected to secondary transformation, the parity of the number of all odd transformation coefficients in the target transformation coefficient matrix (or the parity of the number of all even transform coefficients) is adjusted to an even number. If the current block is not to be transformed twice, the parity of the number of all odd transform coefficients in the target transform coefficient matrix is adjusted to be odd.

For the decoding end, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the third preset condition, then it is judged whether the current block satisfies the fourth preset condition; if the current block does not meet the fourth preset condition condition, the decoder can determine to perform a second inverse transform on the current block. If the current block satisfies the fourth preset condition, the decoding end determines the parity of the number (num_odd) of all odd transform coefficients in the target transform coefficient matrix, and derives secTuflag according to the num_odd, and determines whether to perform the current block based on secTuflag Secondary inverse transformation. For example, if the num_odd is an even number, the value of secTuflag is derived as the first value (such as 0), and at this time, the decoding end determines to perform a second inverse transform on the current block. If the num_odd is an odd number, the derived value of secTuflag is the second value (such as 1), and at this time, the decoder determines not to perform the second inverse transform on the current block.

In the above embodiment, the first preset condition or the third preset condition may include: the prediction mode of the current block is an intra prediction mode; the current block is a luma block; the initial transform coefficient matrix of the current block contains non-zero transform coefficients . The second preset condition or the fourth preset condition may include: the current block does not use the intra prediction filtering mode (ipf_flag=0); the current block does not support the DT division mode (or the current block supports the DT division mode). Optionally, the second preset condition or the fourth preset condition may further include: the width and height of the front block satisfy the first size condition.

Application scenario 5: For the encoding side, if the current block supports secondary transformation, the current block supports AST technology, and the current block meets the first preset condition, it needs to judge whether the current block meets the second preset condition. If the current block meets the The second preset condition is to determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block. If the current block is to be transformed twice, the parity of the number of all odd transform coefficients (or the parity of the number of all even transform coefficients) in the target transform coefficient matrix is adjusted to an even number. If the current block is not to be transformed twice, the parity of the number of all odd transform coefficients in the target transform coefficient matrix is adjusted to be odd.

For the decoder, if the current block supports secondary transformation, the current block supports AST technology, and the current block satisfies the fourth preset condition, then it is judged whether the current block satisfies the fourth preset condition, and if the current block satisfies the fourth preset condition , the decoding end determines the parity of the number (num_odd) of all odd transform coefficients in the target transform coefficient matrix, and derives secTuflag according to the num_odd, and determines whether to perform a second inverse transform on the current block based on secTuflag. For example, if the num_odd is an even number, the value of secTuflag is derived as the first value (such as 0), and at this time, the decoding end determines to perform a second inverse transform on the current block. If the num_odd is an odd number, the derived value of secTuflag is the second value (such as 1), and at this time, the decoder determines not to perform the second inverse transform on the current block.

In the above embodiment, the first preset condition or the third preset condition may include: the prediction mode of the current block is an intra prediction mode; the current block is a luma block; the initial transform coefficient matrix of the current block contains non-zero transform coefficients . The second preset condition or the fourth preset condition may include: the current block uses the intra prediction filtering mode (ipf_flag=1); the current block does not support the DT division mode (or the current block supports the DT division mode). Optionally, the second preset condition or the fourth preset condition may further include: the width and height of the front block satisfy the first size condition.

Application scenario 6: For the encoding side, if the current block is transformed twice, the parity of the number of all odd transform coefficients in the target transform coefficient matrix is adjusted to be odd. If the current block is not to be transformed twice, the parity of the number of all odd transform coefficients in the target transform coefficient matrix is adjusted to be even. For the decoding end, determine the parity of the number (num_odd) of all odd transform coefficients in the target transform coefficient matrix, and derive secTuflag according to the num_odd, and determine whether to perform a second inverse transform on the current block based on secTuflag. For example, if num_odd is an odd number, the value of secTuflag is derived as the first value (such as 0), and it is determined to perform a second inverse transformation on the current block. If num_odd is an even number, then derive the value of secTuflag as the second value (such as 1), and determine not to perform the second inverse transform on the current block.

For determining whether to perform secondary transformation on the current block at the encoding end, refer to application scenario 4 or application scenario 5, and to determine whether to perform secondary transformation on the current block at the decoding end, please refer to application scenario 4 or application scenario 5, which will not be repeated here.

Application Scenario 7: On the basis of Application Scenario 4-Application Scenario 6, for the encoding side, if it is determined that the current block needs to be re-transformed, then the current block is re-transformed based on the size constraints of the current block. For example, if the width and height of the current block satisfy the second size condition, a second transform is performed on the transform coefficients of the first specified area (such as 8*8 area) located in the upper left corner in the initial transform coefficient matrix. If the width and height of the current block do not satisfy the second size condition, perform secondary transformation on the transformation coefficients of the second specified area (eg 4*4 area) located in the upper left corner in the initial transformation coefficient matrix.

Exemplarily, for the decoding end, if it is determined that the current block needs to be subjected to the second inverse transformation, then the current block may be subjected to the second inverse transformation based on the size restriction condition of the current block. For example, if the width and height of the current block satisfy the second size condition, a second inverse transform may be performed on the transform coefficients of the first specified area (eg, 8*8 area) located in the upper left corner of the target transform coefficient matrix. If the width and height of the current block do not satisfy the second size condition, a second inverse transform may be performed on the transform coefficients of the second specified area (eg, 4*4 area) located in the upper left corner of the target transform coefficient matrix.

Embodiment 9: In the foregoing embodiments 1 to 8, the intra prediction filter (Intra Prediction Filter, IPF) is involved, and the intra prediction filter technology will be described below. Intra-frame prediction filtering can use 1 bit to indicate whether intra-frame prediction filtering is allowed for the current block, and it can be determined in the RDO stage whether intra-frame prediction filtering needs to be used for the current block. In addition, the intra prediction filtering technology needs to select reference pixels according to the intra prediction mode of the current block, which can be divided into the following three cases.

Case 1: For DC mode, Plane mode and Bilinear mode, use a column of reconstructed pixels on the left and a row of reconstructed pixels on the upper side as reference pixels, and use a 3-tap filter. For example, the following formula can be used for intra prediction filtering:

P'(x, y) = f(x)·P(-1, y)+f(y)·P(x,-1)+(1-f(x)-f(y))·P( x,y)

In the above formula, 0≤x, y<N, P'(x, y) is the filtered intra-frame prediction value, P(x, y) is the intra-frame prediction value obtained by the conventional intra-frame prediction mode, f( x) and f(y) are filter coefficients, and N is the size of the current block.

Case 2: For the vertical prediction mode (such as mode 3 to mode 18) in HPM, use a column of reconstructed pixels on the left as reference pixels, and use a 2-tap filter. For example, the following formula can be used for intra prediction filtering:

P'(x,y)=f(x)·P(-1,y)+(1-f(x))·P(x,y)

In the above formula, x<N, P'(x, y) can be the filtered intra-frame prediction value, P(x, y) can be the intra-frame prediction value obtained by the conventional intra-frame prediction mode, f(x) and f(y) may be filter coefficients, and N is the size of the current block.

Case 3: For the horizontal prediction mode in HPM (such as mode 19 to mode 32), use the upper row of reconstructed pixels as reference pixels, and use a 2-tap filter. For example, the following formula can be used for intra prediction filtering:

P'(x,y)=f(y)·P(x,-1)+(1-f(y))·P(x,y)

In the above formula, y<N, P'(x, y) can be the filtered intra-frame prediction value, P(x, y) can be the intra-frame prediction value obtained by the conventional intra-frame prediction mode, f(x) and f(y) may be filter coefficients, and N is the size of the current block.

Exemplarily, for case 1, case 2 and case 3, in order to avoid floating-point operations, the filter coefficients can also be fixed-point, such as the intra prediction filter formula for case 1, which can be replaced with the following formula, in the following formula, F(x) and F(y) are the coefficients after fixed-point conversion, that is, f(x) and f(y) are converted into F(x) and F(y). Case 2 is similar to case 3, and will not be repeated here. .

p'(x)=(F(x)·p(-1,y)+F(y)·p(x,-1)+(64-F(x)-F(y))·p(x )+32)>>6

Exemplarily, F(x) and F(y) may be filter coefficients (that is, filter coefficients after amplification), as shown in Table 3, which shows an example of filter coefficients, which are related to the block size and The distance from the reference pixel is all relevant. In Table 3, the prediction block size represents the width or height of the current block, and the distance from the reference pixel represents the prediction distance. In Table 3, the maximum prediction distance is set to 10. In practical applications, the maximum prediction distance can be larger. For This is not limited.

table 3

For example, for the pixel position A of the first row and the third column of the current block, assuming that the reference pixel is a column of reconstructed pixels on the left, the distance between the pixel position A and the reference pixel is 3, assuming that the size of the current block is 4*4, Then the filter coefficient can be 2, assuming the size of the current block is 8*8, the filter coefficient can be 14, assuming the size of the current block is 16*16, the filter coefficient can be 19, assuming the size of the current block is 32*32 , the filter coefficient can be 21, and so on. For other pixel positions, the method of determining the filter coefficient is similar to the method of determining the filter coefficient of pixel position A, which will not be repeated here.

Embodiment 10: In the above-mentioned Embodiment 1-Embodiment 8, the encoding end performs secondary transformation on the transformation coefficient of the 4*4 area located in the upper left corner, or the decoding end performs transformation on the 4*4 area located in the upper left corner Coefficients undergo a second inverse transformation. From the encoding side, the transformation coefficient is obtained after (DCT2, DCT2) transformation, and then the 4*4 transformation coefficient in the upper left corner is multiplied by the secondary transformation matrix to obtain the new transformation coefficient, and then quantized and entropy coding. From the perspective of the decoder, after inverse quantization of the transformation coefficients obtained from entropy decoding of the code stream, a second inverse transformation is performed on the 4*4 area in the upper left corner to obtain new transformation coefficients, and then the entire transformation block is used (DCT2, DCT2) is inversely transformed to obtain residual coefficients.

In a possible implementation manner, the specific inverse transformation process at the decoding end of the secondary transformation is as follows:

Case 1, if the value of the width or height of the current block is greater than 4, and the value of the secondary transformation enable flag is equal to 1 (the value of the secondary transformation enable flag is equal to 1, indicating that the sequence level or frame level allows secondary transformation to be enabled), then Perform the following operations on the coefficient matrix: First, obtain the 4*4 coefficient matrix C in the upper left corner from the transformation coefficient matrix: c _ij =coeff _ij , i=0~3, j=0~3, in the above formula, c _ij is The elements of the matrix C, coeff _ij are the elements of the transformation coefficient matrix.

Exemplarily, if the value of the intra prediction mode is 0~2, or 13~32, or 44~65, and the reference sample on the left outside the current block is "available", then: c _ij =Clip3(-32768, 32767, (p _ij +2 ⁶ )>>7), i=0~3, j=0~3. p _ij is the element of 4*4 matrix P, and the calculation method of matrix P is as follows: P=C×S ₄ , S ₄ is the 4*4 inverse transformation matrix, Clip3(-32768, 32767, (p _ij +2 ⁶ ) >>7) means that if the value of c _ij is between -32768 and 32767, then the value of c _ij is (p _ij +2 ⁶ )>>7, if the value of c _ij is less than -32768, then the value of c _ij is -32768, if the value of c _ij is greater than 32767, then the value of c _ij is 32767.

Exemplarily, if the value of the intra prediction mode is 0 to 23, or 34 to 57, and the reference sample on the outside of the current block is "available", then: c _ij =Clip3(-32768,32767,(q _ij +2 ⁶ )>>7), i=0～3, j=0～3. q _ij is the element of 4×4 matrix Q, and the calculation method of matrix Q is as follows: Q=S ₄ ^T ×C, S ₄ ^T is the transposition matrix of S ₄ , and S ₄ is the 4*4 inverse transformation matrix.

Then, modify the values of the elements of the transformation coefficient matrix according to the matrix C: coeff _ij = _cij , i=0~3, j=0~3.

Then, an initial inverse transformation is performed on the matrix coeff.

Case 2, if the values of the width and height of the current block are both equal to 4, and the value of the secondary transformation enabling flag is equal to 1 (the value of the secondary transformation enabling flag is equal to 1, indicating that the secondary transformation is allowed to be enabled at the sequence level or frame level), Then perform the following operations on the coefficient matrix: First, perform the following vertical inverse transformation on the transformation coefficient matrix to obtain the matrix K: k _ij =Clip3(-32768,32767,(v _ij +2 ⁴ )>>5), i=0～ M ₁ -1, j=0 to M ₂ -1, v _ij is an element of matrix V, and _kij is an element of matrix K. The calculation of the matrix V is as follows: V=D ₄ ^T ×CoeffMatrix, D ₄ ^T is the transpose matrix of the inverse transformation matrix D ₄ , and CoeffMatrix is the transformation coefficient.

Then, perform the following horizontal inverse transformation on matrix K to obtain matrix H: h _ij =Clip3(-MaxValue-1,MaxValue,(w _ij +2 ^shift1-1 )>>shift1), i=0～M ₁ -1, j=0 to M ₂ -1, w _i,j are elements of matrix W, and h _ij are elements of matrix H. The value of MaxValue is (1<<BitDepth)-1, shift1 is equal to 22-BitDepth, the matrix W is: W=K×D ₄ , and D ₄ is an inverse transformation matrix.

Exemplarily, for a 4*4 block, the initial inverse transformation does not need to be performed after the inverse transformation of the secondary transformation is performed.

Embodiment 11: In the above-mentioned Embodiment 1-Embodiment 8, the encoding end performs secondary transformation on the transformation coefficient of the 8*8 area located in the upper left corner, or the decoding end performs transformation on the 8*8 area located in the upper left corner Coefficients undergo a second inverse transformation. From the encoding side, the transformation coefficient is obtained after (DCT2, DCT2) transformation, and then the 8*8 transformation coefficient in the upper left corner is multiplied by the secondary transformation matrix to obtain the new transformation coefficient, and then quantized and entropy coding. From the decoding side, after inverse quantization of the transformation coefficients obtained from the entropy decoding of the code stream, a second inverse transformation is performed on the 8*8 area in the upper left corner to obtain new transformation coefficients, and then the entire transformation block is used (DCT2, DCT2) is inversely transformed to obtain residual coefficients.

In a possible implementation manner, the specific inverse transformation process at the decoding end of the secondary transformation is as follows:

Case 1, if the width or height of the current block is greater than 8, and the value of the secondary transformation enable flag is equal to 1 (the value of the secondary transformation enable flag is equal to 1, indicating that the sequence level or frame level allows secondary transformation to be enabled), then Perform the following operations on the coefficient matrix: First, obtain the 8*8 coefficient matrix C in the upper left corner from the transformation coefficient matrix: c _ij =coeff _ij , i=0~7, j=0~7, in the above formula, c _ij is The elements of the matrix C, coeff _ij are the elements of the transformation coefficient matrix.

Exemplarily, if the value of the intra prediction mode is 0~2, or 13~32, or 44~65, and the reference sample on the left outside the current block is "available", then: c _ij =Clip3(-32768, 32767, (p _ij +2 ⁶ )>>7), i=0~7, j=0~7. p _ij is the element of matrix P of 8*8, the calculation method of matrix P is as follows: P=C×S ₈ , S ₈ is the inverse transformation matrix of 8*8, Clip3(-32768, 32767, (p _ij +2 ⁶ ) >>7) means that if the value of c _ij is between -32768 and 32767, then the value of c _ij is (p _ij +2 ⁶ )>>7, if the value of c _ij is less than -32768, then the value of c _ij is -32768, if the value of c _ij is greater than 32767, then the value of c _ij is 32767.

Exemplarily, if the value of the intra prediction mode is 0 to 23, or 34 to 57, and the reference sample on the outside of the current block is "available", then: c _ij =Clip3(-32768,32767,(q _ij +2 ⁶ )>>7), i=0～7, j=0～7. q _ij is an element of the 8×8 matrix Q, and the calculation method of the matrix Q is as follows: Q=S ₈ ^T ×C, S ₈ ^T is the transposition matrix of S ₈ , and S ₈ is the 8*8 inverse transformation matrix.

Then, modify the values of the elements of the transformation coefficient matrix according to the matrix C: coeff _ij =c _ij , i=0-7, j=0-7.

Then, an initial inverse transformation is performed on the matrix coeff.

Case 2, if the values of the width and height of the current block are both equal to 8, and the value of the secondary transformation enabling flag is equal to 1 (the value of the secondary transformation enabling flag is equal to 1, indicating that the secondary transformation is allowed to be enabled at the sequence level or frame level), Then perform the following operations on the coefficient matrix: First, perform the following vertical inverse transformation on the transformation coefficient matrix to obtain the matrix K: k _ij =Clip3(-32768,32767,(v _ij +2 ⁴ )>>5), i=0～ M ₁ -1, j=0 to M ₂ -1, v _ij is an element of matrix V, and _kij is an element of matrix K. The calculation of the matrix V is as follows: V=D ₈ ^T ×CoeffMatrix, D ₈ ^T is the transpose matrix of the inverse transformation matrix D ₈ , and CoeffMatrix is the transformation coefficient.

Then, perform the following horizontal inverse transformation on matrix K to obtain matrix H: h _ij =Clip3(-MaxValue-1,MaxValue,(w _ij +2 ^shift1-1 )>>shift1), i=0～M ₁ -1, j=0 to M ₂ -1, w _i,j are elements of matrix W, and h _ij are elements of matrix H. The value of MaxValue is (1<<BitDepth)-1, shift1 is equal to 22-BitDepth, the matrix W is: W=K×D ₈ , and D ₈ is an inverse transformation matrix.

Certainly, the above-mentioned embodiment 1 to embodiment 11 can be implemented independently or in combination. For example, embodiment 1 and embodiment 2 can be realized in combination, embodiment 3 and embodiment 4 can be realized in combination, embodiment 5 and embodiment 6 can be realized in combination, embodiment 1 and embodiment 3 can be realized in combination, embodiment 1 and embodiment Embodiment 5 can be realized in combination, embodiment 2 and embodiment 4 can be realized in combination, embodiment 2 and embodiment 6 can be realized in combination, embodiment 3 and embodiment 7 can be realized in combination, embodiment 4 and embodiment 7 can be realized in combination , Embodiment 5 and Embodiment 8 can be implemented in combination, and Embodiment 6 and Embodiment 8 can be implemented in combination. Of course, the above are just a few examples, and the combinations among these embodiments are not limited.

Based on the same application concept as the above method, the embodiment of the present application also proposes an encoding device, which is applied to the encoding end, as shown in Figure 5A, which is a structural diagram of the device, and the device includes:

An acquisition module 511, configured to acquire a residual coefficient matrix corresponding to the current block;

A transformation module 512, configured to perform an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix;

Determining module 513, used to determine whether the current block meets the second preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the first preset condition; If the block satisfies the second preset condition, determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block;

Wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

Exemplarily, the determination module 513 is further configured to: if performing secondary transformation on the current block, perform secondary transformation on the transformation coefficients of the designated area located in the upper left corner in the initial transformation coefficient matrix to obtain the target transformation A coefficient matrix; if no secondary transformation is performed on the current block, the initial transformation coefficient matrix is determined as a target transformation coefficient matrix;

Encoding is performed according to the target transformation coefficient matrix to obtain an encoded bit stream of the current block.

The determination module 513 is specifically used for performing secondary transformation on the transformation coefficients of the specified area located in the upper left corner in the initial transformation coefficient matrix: if the width and height of the current block meet the second size condition, then based on the first A secondary transformation matrix of a specified size, which performs a secondary transformation on the transformation coefficients of the first specified area in the upper left corner of the initial transformation coefficient matrix; if the width and height of the current block do not meet the second size condition, then based on A secondary transformation matrix of a second specified size, performing secondary transformation on the transformation coefficients of the second specified area located in the upper left corner of the initial transformation coefficient matrix;

Wherein, the size of the first specified area is a first specified size, the size of the second specified area is a second specified size, and the first specified size is larger than the second specified size.

The determining module 513 is further configured to: add first indication information to the coded bit stream if the current block is to be transformed twice; wherein, the first indication information is used to indicate that the current block perform a secondary transformation; or,

If the second transformation is not performed on the current block, second indication information is added to the coded bit stream; wherein the second indication information is used to indicate that the current block is not to be transformed twice.

The determining module 513 performs encoding according to the target transformation coefficient matrix, and when obtaining the coded bit stream of the current block, is specifically used to: determine the target transformation coefficient matrix of the target transformation coefficient matrix according to the transformation coefficients in the target transformation coefficient matrix Features; according to the target feature and the target value of the preset flag bit of the current block, determine whether to adjust the target transform coefficient matrix; when the target value of the preset flag bit is the first value , used to indicate that a second transformation is performed on the current block, and when the target value of the preset flag bit is the second value, it is used to indicate that the current block is not to be transformed twice;

If the target transform coefficient matrix is not adjusted, encode the target transform coefficient matrix to obtain the coded bit stream of the current block; if the target transform coefficient matrix is adjusted, the target transform coefficient matrix The transformation coefficients in the matrix are adjusted so that the target characteristics of the adjusted target transformation coefficient matrix match the target value of the preset flag bit, and the adjusted target transformation coefficient matrix is encoded to obtain the current The encoded bitstream of the block.

Based on the same application concept as the above method, the embodiment of the present application also proposes a decoding device, which is applied to the decoding end, as shown in Figure 5B, which is a structural diagram of the device, and the device includes:

An acquisition module 521, configured to acquire the coded bitstream of the current block;

The determination module 522 is used to determine whether the current block meets the fourth preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the third preset condition; block satisfies the fourth preset condition, then determine whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix; wherein, the third preset condition includes: the The prediction mode of the current block is an intra prediction mode; the fourth preset condition includes: the current block uses an intra prediction filtering mode, or does not use an intra prediction filtering mode.

When the determining module 522 determines whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream, it is specifically used to: parse out the indication information from the coded bit stream; if the indication information is The first indication information is to determine to perform a second inverse transformation on the current block; wherein, the first indication information is used to indicate to perform a second transformation on the current block; or, if the indication information is a second indication information, it is determined not to perform secondary inverse transformation on the current block; wherein, the second indication information is used to indicate not to perform secondary transformation on the current block.

When the determining module 522 determines whether to perform a second inverse transform on the current block according to the target transformation coefficient matrix in the coded bitstream, it is specifically used to: parse out the target transformation coefficient matrix from the coded bitstream;

determining target features of the target transform coefficient matrix according to transform coefficients in the target transform coefficient matrix;

Determine the target value of the preset flag bit of the current block according to the target feature; wherein, when the target value of the preset flag bit is the first value, it is used to indicate that the current block is performed twice Inverse transformation, when the target value of the preset flag bit is the second value, it is used to indicate that the current block is not to be subjected to a second inverse transformation;

Determine whether to perform secondary inverse transformation on the current block according to the target value of the preset flag bit.

The determination module 522 is also used to: parse out the target transformation coefficient matrix from the coded bit stream; if the second inverse transformation is performed on the current block, the specified region located in the upper left corner of the target transformation coefficient matrix The transformation coefficients of the transformation coefficients are subjected to a second inverse transformation to obtain an initial transformation coefficient matrix; if the current block is not subjected to a second inverse transformation, the target transformation coefficient matrix is determined as an initial transformation coefficient matrix; the initial transformation coefficient matrix Perform initial inverse transformation to obtain a residual coefficient matrix corresponding to the current block; determine a reconstruction value of the current block according to the residual coefficient matrix.

The determining module 522 is specifically used to perform a second inverse transform on the transform coefficients of the specified region located in the upper left corner in the target transform coefficient matrix: if the width and height of the current block meet the second size condition, then based on the second size condition A secondary inverse transformation matrix of a specified size, performing a secondary inverse transformation on the transformation coefficients of the first specified area in the upper left corner of the target transformation coefficient matrix; if the width and height of the current block do not meet the second size condition , then based on the secondary inverse transformation matrix of the second specified size, perform a secondary inverse transformation on the transformation coefficients of the second specified region located in the upper left corner of the target transformation coefficient matrix; wherein, the size of the first specified region is A first specified size, the size of the second specified area is a second specified size, and the first specified size is larger than the second specified size.

The encoding end device provided by the embodiment of the present application (for example, the encoding end device may be a video encoder), in terms of hardware level, its hardware architecture schematic diagram can refer to FIG. 6A for details. Including: a processor 611 and a machine-readable storage medium 612, the machine-readable storage medium 612 stores machine-executable instructions that can be executed by the processor 611; the processor 611 is configured to execute the machine-executable instructions, In order to realize the method disclosed in the above example of the present application. For example, the processor is used to execute machine-executable instructions to implement the following steps: obtaining the residual coefficient matrix corresponding to the current block;

performing an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix;

If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the first preset condition, then determine whether the current block satisfies the second preset condition; if the current block satisfies the second preset condition condition, then determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block;

Wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

The decoding end device provided by the embodiment of the present application (for example, the decoding end device may be a video decoder), in terms of hardware level, its hardware architecture schematic diagram can be specifically referred to as shown in FIG. 6B . Including: a processor 621 and a machine-readable storage medium 622, the machine-readable storage medium 622 stores machine-executable instructions that can be executed by the processor 621; the processor 621 is used to execute the machine-executable instructions, In order to realize the method disclosed in the above example of the present application. For example, the processor is configured to execute machine-executable instructions to implement the steps of: obtaining an encoded bitstream of the current block;

If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block meets the third preset condition, then determine whether the current block meets the fourth preset condition; if the current block meets the fourth preset condition condition, then determine whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix;

Wherein, the third preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the fourth preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. Predictive filtering mode.

Based on the same application concept as the above method, an embodiment of the present application further provides a camera device, and the camera device may include the encoding device and/or the decoding device in any of the above embodiments.

Based on the same application idea as the above-mentioned method, the embodiment of the present application also provides a machine-readable storage medium, on which several computer instructions are stored, and when the computer instructions are executed by a processor, the present invention can be realized. Apply the codec method disclosed in the above example. Wherein, the above-mentioned machine-readable storage medium may be any electronic, magnetic, optical or other physical storage device, which may contain or store information, such as executable instructions, data, and so on. For example, the machine-readable storage medium can be: RAM (Radom Access Memory, random access memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid-state hard disk, any type of storage disk (such as CD, DVD, etc.), or similar storage media, or a combination of them.

The systems, devices, modules or units described in the above embodiments may be realized by computer chips or entities, or by products with certain functions. A typical implementation is a computer, which may be in the form of a personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media player, navigation device, e-mail device, tablet computer, A wearable device or a combination of any of these devices.

For the convenience of description, when describing the above devices, functions are divided into various units and described separately. Of course, when implementing the present application, the functions of each unit can be implemented in one or more pieces of software and/or hardware.

The above descriptions are only examples of the present application, and are not intended to limit the present application. For those skilled in the art, various modifications and changes may occur in this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims (16)

1. An encoding method, characterized in that being applied to an encoding end, the method comprises: Obtain the residual coefficient matrix corresponding to the current block; performing an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix; If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the first preset condition, then determine whether the current block satisfies the second preset condition; if the current block satisfies the second preset condition condition, then determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block; The transformation coefficients of the designated area of the specified area are subjected to secondary transformation to obtain the target transformation coefficient matrix; if the current block is not subjected to secondary transformation, the initial transformation coefficient matrix is determined as the target transformation coefficient matrix; according to the target transformation coefficient matrix to obtain the coded bit stream of the current block; Wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. predictive filter mode; Wherein, if the current block does not meet the first preset condition, the second transformation of the current block is prohibited. 2. The method according to claim 1, wherein the second preset condition comprises: The width and height of the current block satisfy the first size condition, and the current block does not support the DT division mode. 3. The method of claim 1, wherein, performing a secondary transformation on the transformation coefficients in the specified area located in the upper left corner in the initial transformation coefficient matrix, including: If the width and height of the current block satisfy the second size condition, perform a secondary transformation on the transformation coefficients of the first specified area located in the upper left corner of the initial transformation coefficient matrix based on the secondary transformation matrix of the first specified size ; If the width and height of the current block do not meet the second size condition, based on the secondary transformation matrix of the second specified size, the transformation coefficients of the second specified area located in the upper left corner of the initial transformation coefficient matrix are performed twice transform; Wherein, the size of the first specified area is a first specified size, the size of the second specified area is a second specified size, and the first specified size is larger than the second specified size. 4. The method according to claim 1, wherein said encoding according to said target transformation coefficient matrix, and after obtaining the encoded bit stream of said current block, said method further comprises: If performing secondary transformation on the current block, adding first indication information to the coded bit stream; wherein the first indication information is used to indicate performing secondary transformation on the current block; or, If the second transformation is not performed on the current block, second indication information is added to the coded bit stream; wherein the second indication information is used to indicate that the current block is not to be transformed twice. 5. The method of claim 1, wherein, The encoding according to the target transformation coefficient matrix to obtain the encoded bit stream of the current block includes: determining target features of the target transform coefficient matrix according to transform coefficients in the target transform coefficient matrix; According to the target feature and the target value of the preset flag bit of the current block, determine whether to adjust the target transformation coefficient matrix; when the target value of the preset flag bit is the first value, use When instructing to perform secondary transformation on the current block, when the target value of the preset flag bit is the second value, it is used to indicate not to perform secondary transformation on the current block; If the target transform coefficient matrix is not adjusted, encode the target transform coefficient matrix to obtain the coded bit stream of the current block; if the target transform coefficient matrix is adjusted, the target transform coefficient matrix The transformation coefficients in the matrix are adjusted so that the target characteristics of the adjusted target transformation coefficient matrix match the target value of the preset flag bit, and the adjusted target transformation coefficient matrix is encoded to obtain the current The encoded bitstream of the block. 6. The method of claim 5, wherein, The target feature includes target parity, and determining whether to adjust the target transform coefficient matrix according to the target feature and the target value of the preset flag bit of the current block includes: If the target parity is an odd number, and the target value of the preset flag bit of the current block is the first value, then determine not to adjust the target transform coefficient matrix; or, If the target parity is an odd number, and the target value of the preset flag bit of the current block is the second value, then determine to adjust the target transform coefficient matrix; or, If the target parity is an even number, and the target value of the preset flag bit of the current block is the first value, then determine to adjust the target transform coefficient matrix; or, If the target parity is an even number, and the target value of the preset flag bit of the current block is a second value, then determine not to adjust the target transform coefficient matrix; Wherein, when the target parity is an odd number, it is used to indicate to perform a secondary transformation on the current block; When the target parity is an even number, it is used to indicate not to perform secondary transformation on the current block. 7. The method of claim 5, wherein, The target feature includes target parity, and determining whether to adjust the target transform coefficient matrix according to the target feature and the target value of the preset flag bit of the current block includes: If the target parity is an even number, and the target value of the preset flag bit of the current block is the first value, then determine not to adjust the target transform coefficient matrix; or, If the target parity is an even number, and the target value of the preset flag bit of the current block is the second value, then determine to adjust the target transform coefficient matrix; or, If the target parity is an odd number, and the target value of the preset flag bit of the current block is the first value, then determine to adjust the target transform coefficient matrix; or, If the target parity is an odd number, and the target value of the preset flag bit of the current block is a second value, then determine not to adjust the target transform coefficient matrix; Wherein, when the target parity is an even number, it is used to indicate to perform secondary transformation on the current block; When the target parity is an odd number, it is used to indicate that no secondary transformation is performed on the current block. 8. A decoding method, characterized in that it is applied to a decoding end, and the method comprises: Get the encoded bitstream of the current block; If the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block meets the third preset condition, then determine whether the current block meets the fourth preset condition; if the current block meets the fourth preset condition condition, then determine whether to perform a second inverse transform on the current block according to the indication information in the coded bit stream or the target transform coefficient matrix; parsing a target transform coefficient matrix from the encoded bitstream; If a second inverse transform is performed on the current block, then a second inverse transform is performed on the transform coefficients in the specified area in the upper left corner of the target transform coefficient matrix to obtain an initial transform coefficient matrix; if the current block is not performed secondary inverse transform, the target transform coefficient matrix is determined as the initial transform coefficient matrix; performing an initial inverse transform on the initial transform coefficient matrix to obtain a residual coefficient matrix corresponding to the current block; determining a reconstruction value of the current block according to the residual coefficient matrix; Wherein, the third preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the fourth preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. predictive filter mode; Wherein, if the current block does not satisfy the third preset condition, the second inverse transformation for the current block is prohibited. 9. The method according to claim 8, wherein the fourth preset condition comprises: The width and height of the current block satisfy the first size condition, and the current block does not support the DT division mode. 10. The method of claim 8, wherein, The determining whether to perform secondary inverse transformation on the current block according to the indication information in the coded bit stream includes: parsing the indication information from the coded bit stream; If the indication information is the first indication information, determine to perform a second inverse transformation on the current block; where the first indication information is used to indicate to perform a second transformation on the current block; or, If the indication information is the second indication information, it is determined not to perform the second inverse transformation on the current block; wherein the second indication information is used to indicate not to perform the second transformation on the current block. 11. The method according to claim 8, wherein the determining whether to perform a second inverse transform on the current block according to the target transform coefficient matrix in the coded bitstream comprises: parsing a target transform coefficient matrix from the encoded bitstream; determining target features of the target transform coefficient matrix according to transform coefficients in the target transform coefficient matrix; Determine the target value of the preset flag bit of the current block according to the target feature; wherein, when the target value of the preset flag bit is the first value, it is used to indicate that the current block is performed twice Inverse transformation, when the target value of the preset flag bit is the second value, it is used to indicate that the current block is not to be subjected to a second inverse transformation; Determine whether to perform secondary inverse transformation on the current block according to the target value of the preset flag bit. 12. The method according to claim 11, wherein the target feature comprises target parity, and determining the target value of the preset flag bit of the current block according to the target feature comprises: If the target parity is an odd number, then determine that the target value of the preset flag bit is the first value; If the target parity is an even number, then determine that the target value of the preset flag bit is a second value; The determining whether to perform secondary inverse transformation on the current block according to the target value of the preset flag bit includes: If the target value of the preset flag bit is the first value, then determine to perform a second inverse transform on the current block; If the target value of the preset flag bit is the second value, it is determined not to perform a second inverse transform on the current block. 13. The method according to claim 11, wherein the target feature comprises target parity, and determining the target value of the preset flag bit of the current block according to the target feature comprises: If the target parity is an even number, then determine that the target value of the preset flag bit is the first value; If the target parity is an odd number, then determine that the target value of the preset flag bit is a second value; The determining whether to perform secondary inverse transformation on the current block according to the target value of the preset flag bit includes: If the target value of the preset flag bit is the first value, then determine to perform a second inverse transform on the current block; If the target value of the preset flag bit is the second value, it is determined not to perform a second inverse transform on the current block. 14. The method of claim 8, wherein, Performing a second inverse transformation on the transformation coefficients of the specified region located in the upper left corner in the target transformation coefficient matrix, including: If the width and height of the current block satisfy the second size condition, based on the second inverse transform matrix of the first specified size, the transform coefficients of the first specified area located in the upper left corner of the target transform coefficient matrix are twice performed inverse transform; If the width and height of the current block do not meet the second size condition, then based on the second specified size of the secondary inverse transformation matrix, the transformation coefficient of the second specified region located in the upper left corner of the target transformation coefficient matrix is performed twice. secondary inverse transformation; Wherein, the size of the first specified area is a first specified size, the size of the second specified area is a second specified size, and the first specified size is larger than the second specified size. 15. An encoding device, characterized in that it is applied to an encoding end, and the device comprises: An acquisition module, configured to acquire a residual coefficient matrix corresponding to the current block; A transformation module, configured to perform an initial transformation on the residual coefficient matrix to obtain an initial transformation coefficient matrix; A determining module, configured to determine whether the current block satisfies the second preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the first preset condition; if the current block If the second preset condition is met, then determine whether to perform secondary transformation on the current block according to the rate-distortion cost value corresponding to the current block; performing a secondary transformation on the transformation coefficients of the specified region located in the upper left corner of the matrix to obtain a target transformation coefficient matrix; if the current block is not subjected to secondary transformation, the initial transformation coefficient matrix is determined as the target transformation coefficient matrix; according to Encoding the target transformation coefficient matrix to obtain an encoded bit stream of the current block; Wherein, the first preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the second preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. predictive filter mode; Wherein, if the current block does not meet the first preset condition, the second transformation of the current block is prohibited. 16. A decoding device, characterized in that it is applied to a decoding end, and the device comprises: An acquisition module, configured to acquire the coded bitstream of the current block; A determining module, configured to determine whether the current block satisfies the fourth preset condition if the current block supports secondary transformation, the current block supports the optional secondary transformation AST technology, and the current block satisfies the third preset condition; if the current block Satisfy the fourth preset condition, then determine whether to perform a second inverse transform on the current block according to the indication information in the encoded bit stream or the target transform coefficient matrix; parse out the target transform coefficient matrix from the encoded bit stream; If a second inverse transform is performed on the current block, then a second inverse transform is performed on the transform coefficients in the specified area in the upper left corner of the target transform coefficient matrix to obtain an initial transform coefficient matrix; if the current block is not performed In the second inverse transformation, the target transformation coefficient matrix is determined as an initial transformation coefficient matrix; an initial inverse transformation is performed on the initial transformation coefficient matrix to obtain a residual coefficient matrix corresponding to the current block; according to the residual coefficient matrix a matrix determining reconstruction values for said current block; Wherein, the third preset condition includes: the prediction mode of the current block is an intra-frame prediction mode; the fourth preset condition includes: the current block uses an intra-frame prediction filtering mode, or does not use an intra-frame prediction mode. predictive filter mode; Wherein, if the current block does not satisfy the third preset condition, the second inverse transformation for the current block is prohibited.

Images