CN112104876A - Data compression method and device for performing multi-set coefficient component conversion on prediction residual error - Google Patents

Abstract

The present invention provides a data compression method and device for converting multiple sets of coefficient components to prediction residuals. In the present method and apparatus, for multi-component residual data, according to a predetermined rule and according to its characteristics, one of the predetermined at least three sets of coefficients is selected for component conversion, and the predetermined at least three sets of coefficients are allowed to include identity conversion. That is, a set of coefficients that do not actually do component conversion.

Description

Data compression method and device for converting multiple sets of coefficient components to prediction residuals

technical field

The present invention relates to an encoding and decoding system for lossy or lossless compression of data, in particular to a method and device for encoding and decoding prediction residual data.

Background technique

With the entry into the era of artificial intelligence, big data, cloud-edge computing, and 5G, ultra-high compression ratio and high-quality data compression have become an indispensable technology for data including image and video data and various new forms of data.

A dataset is a collection of data elements (eg: bytes, bits, pixels, pixel components, spatial samples, transform domain coefficients).

When encoding or decoding a data set (referred to as encoding and decoding), the data elements are usually sorted according to predetermined rules, that is, the order is specified, and the encoding and decoding are performed in the order.

For datasets arranged in a certain spatial (one-, two-, or multi-dimensional) shape (for example: a one-dimensional data queue, a two-dimensional data file, a frame of image, a video sequence, a transform domain, a transform block, Multiple transform blocks, a 3D scene, a sequence of continuously changing 3D scenes), especially when two-dimensional or more data sets are encoded (and correspondingly decoded) for data compression, sometimes the data set is divided into several A subset of a predetermined shape and/or size (ie, the number of elements) is called an integer compression unit, and is encoded or decoded one by one in a predetermined order with the integer compression unit as a unit. At any time, the integer compression unit that is being encoded or decoded is called the current integer compression unit. The data element that is being encoded or decoded (sometimes referred to simply as element) is referred to as the current encoded data element or the currently decoded data element, collectively referred to as the current data element, or referred to as the current element for short. An element consists of N components (usually 1≤N≤5), so both datasets and integer compression units also consist of N components.

In the case where the data set is divided into integer compression units, a predetermined rule for sorting is to first sort the integer compression units, and then sort the elements within each integer compression unit.

A predetermined number of integral compression units constitute a compression block. All integral compression units within a compression block usually have one or several characteristics in common.

Coding in data compression usually consists of at least some or all of the following stages:

1) Prediction, mainly including adjacent prediction, string prediction, block prediction, etc., generating prediction values and prediction residuals, referred to as residuals; prediction is also called matching, especially, string prediction is also called string matching, and block prediction is also called block matching. ;

2) Transformation, which is mainly to transform the prediction residual to generate transformation coefficients, referred to as coefficients; when the transformation is an identity transformation, that is, no transformation is actually performed, the coefficients and residuals are exactly equal;

3) Quantization, mainly quantizing the coefficients to generate quantization residuals; when the quantization is identity quantization, that is, when quantization is not actually done, the quantization residuals are equal to the coefficients; when the transformation is identity transformation, the quantization residuals are used to predict The result of quantization of the residual; when the quantization is an identity quantization and the transformation is an identity transformation, the quantization residual is equal to the coefficient and equal to the residual;

4) Entropy coding, which mainly includes entropy coding at least including binarization on the quantized residual to generate a compressed data stream.

The above 2), 3), and 4) are generally collectively referred to as residual coding.

Decoding in data compression usually consists of at least some or all of the following stages:

1) Entropy decoding, mainly to parse the compressed data stream and entropy decoding at least including inverse binarization to generate quantized residuals;

2) Inverse quantization, mainly inverse quantization of the quantization residuals to generate reconstruction coefficients; when quantization is identity quantization, that is, no quantization is actually performed, inverse quantization is also identity inverse quantization, that is, inverse quantization is not actually performed, so The reconstruction coefficient is equal to the quantization residual;

3) Inverse transformation, mainly inverse transformation of the reconstruction coefficient, to generate a reconstruction residual; when the transformation is an identity transformation, that is, no transformation is actually performed, the inverse transformation is also an identity transformation, that is, no inverse transformation is actually performed, so The reconstruction residual is exactly equal to the reconstruction coefficient; when the quantization is identity quantization and the transform is an identity transform, the inverse quantization is also the identity inverse quantization and the inverse transform is also the identity inverse transform, so the reconstruction residual is equal to the reconstruction coefficient Also equal to the quantized residual;

4) Prediction compensation, which mainly includes adjacent prediction compensation, string prediction compensation, block prediction compensation, etc. The prediction value and reconstruction data are also called reconstruction data or restoration data.

The above 1), 2), and 3) are generally collectively referred to as residual decoding.

The prediction residuals, transform coefficients, and quantized residuals in each stage of encoding and the quantized residuals, reconstruction coefficients, and reconstruction residuals in each stage of decoding are collectively referred to as residuals or residual data. Generally, it is determined by context whether the residual is a prediction residual, a transform coefficient, a quantization residual, a reconstruction coefficient or a reconstruction residual. If it cannot be determined from the context, the residuals refer to prediction residuals and/or transform coefficients and/or quantization residuals and/or reconstruction coefficients and/or reconstruction residuals.

The input data, original data, residual data and various intermediate data involved in each stage of encoding and decoding are called data samples or samples for short.

In the case of multi-component data, that is, N>1 data, there is usually a strong correlation between the components of the residual data. A common method in data compression is to use component space transformation (referred to as component transformation) to convert data from a component space with strong correlation between components to another component space with weak correlation between components, so as to improve the subsequent performance. Coding efficiency of entropy coding.

An example of a component transformation positive operation for residual data is:

Component 1 after conversion = (Component 1 before conversion + Component 2 before conversion)/2;

Component 2 after conversion = (Component 1 before conversion - Component 2 before conversion)/2;

The corresponding inverse component transformation operation is:

pre-conversion component 1 = (post-conversion component 1 + post-conversion component 2);

pre-conversion component 2 = (post-conversion component 1 - post-conversion component 2);

In the prior art, a single fixed coefficient is usually used for component transformation of multi-component residual data, and data that cannot adapt to various new forms have various correlations, and even different parts of the same data set have different correlations. new features of sex.

SUMMARY OF THE INVENTION

In order to solve this problem in the prior art, the present invention provides a data compression method and device for performing component transformation on multi-component residual data using multiple sets of coefficients. In the present method and apparatus, for multi-component residual data, according to a predetermined rule and according to its characteristics, one of the predetermined at least three sets of coefficients is selected for component conversion, and the predetermined at least three sets of coefficients are allowed to include identity conversion. That is, a set of coefficients that do not actually do component conversion.

According to one aspect of the present invention, there is provided a multi-component data encoding method or device, comprising at least steps or modules for completing the following functions and operations:

1) Analyze the characteristics of the current multi-component residual data, and according to a predetermined rule, select one of the predetermined sets of coefficients for component conversion as the selected coefficient of the current encoding;

2) at least use the selected coefficients to perform a component transformation forward operation on the current residual;

3) Write the result of the current encoding into the compressed data code stream, and the compressed data code stream includes at least part or all of the information called selected information which is required to indicate which set of coefficients the selected coefficient is.

Fig. 1(a) is a schematic diagram of the encoding method or apparatus of the present invention.

From a first perspective, the present invention provides an encoding method for compressing multi-component data, which is characterized by at least comprising the following steps:

1) Analyze the characteristics of the current multi-component residual data, and according to a predetermined rule, select one of the predetermined sets of coefficients for component conversion as the selected coefficient of the current encoding;

2) at least use the selected coefficients to perform a component transformation forward operation on the current residual;

3) Write the result of the current encoding into the compressed data code stream, and the compressed data code stream includes at least part or all of the information called selected information which is required to indicate which set of coefficients the selected coefficient is.

From a second perspective, the present invention provides an encoding device for compressing multi-component data, which is characterized by at least including the following modules:

Coefficient selection module: analyzes the characteristics of the current multi-component residual data, and selects one of the predetermined multiple sets of coefficients for component conversion as the selected coefficient of the current encoding according to a predetermined rule;

Component transformation forward operation module: use at least the selected coefficient to perform a component transformation forward operation on the current residual;

Code stream generation module: write the result of the current encoding into the compressed data code stream, the compressed data code stream includes at least part or all of the information called selected information which is required to indicate which set of coefficients the selected coefficients are.

According to another aspect of the present invention, there is provided a multi-component data decoding method or device, comprising at least steps or modules for completing the following functions and operations:

1) Parse the compressed data stream, and obtain at least part or all of the information called selection information that indicates which set of coefficients in the predetermined sets of coefficients are selected as the selected coefficients to perform component conversion on the current residual;

2) At least according to the information and/or the parameters involved in the predetermined decoding and/or the variables involved in the predetermined decoding, select one of the predetermined sets of coefficients as the selected coefficient for the current decoding;

3) at least use the selected coefficients to perform an inverse component transformation operation on the residual to obtain restored or reconstructed or reconstructed multi-component residual data.

FIG. 1(b) is a schematic diagram of the decoding method or apparatus of the present invention.

From a third angle, the present invention provides a decoding method for compressing multi-component data, which is characterized by at least comprising the following steps:

1) Parse the compressed data stream, and obtain at least part or all of the information called selection information that indicates which set of coefficients in the predetermined sets of coefficients are selected as the selected coefficients to perform component conversion on the current residual;

2) At least according to the information and/or the parameters involved in the predetermined decoding and/or the variables involved in the predetermined decoding, select one of the predetermined sets of coefficients as the selected coefficient for the current decoding;

3) at least use the selected coefficients to perform an inverse component transformation operation on the residual to obtain restored or reconstructed or reconstructed multi-component residual data.

From a fourth perspective, the present invention provides a decoding device for compressing multi-component data, which is characterized by at least including the following modules:

Code stream parsing module: parses the compressed data stream, and obtains at least some or all of the information, called selected information, which indicates which set of coefficients among the predetermined sets of coefficients is selected as the selected coefficient to perform component conversion on the current residual. ;

Coefficient selection module: selects one of the predetermined sets of coefficients as the selected coefficient for current decoding at least according to the information and/or the parameters involved in the predetermined decoding and/or the variables involved in the predetermined decoding;

Component transformation inverse operation module: at least use the selected coefficients to perform an inverse component transformation operation on the residual to obtain restored or reconstructed or reconstructed multi-component residual data.

The present invention is applicable to encoding and decoding for lossy compression of data, and the present invention is also applicable to encoding and decoding for lossless compression of data. The present invention is applicable to encoding and decoding of one-dimensional data such as character string data or byte string data or one-dimensional graphics or fractal graphics, and is also applicable to encoding and decoding of two-dimensional or more data such as image or video data.

In the present invention, the data involved in data compression includes one or a combination of the following types of data:

1) one-dimensional data; 2) two-dimensional data; 3) multi-dimensional data; 4) graphics; 5) fractal graphics; 6) images; 7) sequences of images; 8) videos; 9) audios; 10) documents; 11 ) bytes; 12) bits; 13) pixels; 14) three-dimensional scenes; 15) sequences of continuously changing three-dimensional scenes; 16) virtual reality scenes; 17) continuously changing sequences of virtual reality scenes; 18) pixel form 19) Transform domain data of the image; 20) A collection of two-dimensional or more bytes; 21) A collection of two-dimensional or more bits; 22) A collection of pixels; 23) A collection of single-component pixels; 24) A set of three-component pixels (R, G, B, A); 25) A set of three-component pixels (Y, U, V); 26) A set of three-component pixels (Y, Cb, Cr); 27) Three set of component pixels (Y, Cg, Co); 28) set of four-component pixels (C, M, Y, K); 29) set of four-component pixels (R, G, B, A); 30) set of four-component pixels set of pixels (Y, U, V, A); 31) set of four-component pixels (Y, Cb, Cr, A); 32) set of four-component pixels (Y, Cg, Co, A).

In the present invention, in the case where the original data is an image, a sequence of images, a video, etc., the integral compression unit is an encoding area or a decoding area of the image or sequence, including the following situations: a sub-image, a slice, a slice of an image tile, tile group, brick brick, macroblock, largest coding unit LCU, coding tree unit CTU, coding unit CU, sub-region of CU, sub-coding unit SubCU, prediction block, prediction unit PU, sub-region of PU , sub-prediction unit SubPU, transform block, transform unit TU, sub-region of TU, sub-transform unit SubTU. In this case, a compressed block is a predetermined number of coding regions or a predetermined number of decoding regions in a picture or sequence, including the following situations: a sequence, a sequence of pictures, a picture, a sub-picture of a picture, a slice, slice tile, slice group tile group, brick brick, largest coding unit LCU, coding tree unit CTU, coding unit CU, transform block, transform unit TU, one or several predetermined number of slices or slice tiles or tile group or brick brick or largest coding unit LCU or coding tree unit CTU or coding unit CU or transform block or transform unit TU.

The technical features of the present invention are described above through some specific specific examples. Those skilled in the art can easily understand other advantages and functions of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

Description of drawings

Fig. 1(a) is a schematic diagram of the encoding method or apparatus of the present invention.

FIG. 1(b) is a schematic diagram of the decoding method or apparatus of the present invention.

Detailed ways

The following are further implementation details or variants of the invention.

Implementation or Variation 1

In the encoding method or apparatus or decoding method or apparatus, the multiple sets of coefficients include at least two sets of coefficients of non-identity transformation.

Implementation or Variation 2

In the encoding method or device or decoding method or device, there is information called integral compression unit component conversion selection information in the compressed data code stream, which indicates which set of coefficients or which coefficients of the multiple sets of coefficients are used by an integral compression unit. Some or all of the information required for the set of coefficients.

Implementation or Variation 3

In the encoding method or apparatus or decoding method or apparatus, component conversion is performed on M (2≤M<N) components among the N (N≥3) components, and component conversion is not performed on the remaining N-M components.

Implementation or Variation 4

和

转换为分量

和

的线性转换G，具有I套系数。In the encoding method or device or decoding method or device, component conversion is performed on 2 of the N (N≥3) components, and component conversion is not performed on the remaining N-2 components; the encoding method or device The component transformation forward operation in is a linear transformation F that transforms components w and x into components y and z, with I (I ≥ 3) sets of coefficients; the component transformation inverse operation in the decoding method or apparatus is the component

and

convert to component

and

The linear transformation G of , with I set of coefficients.

，

，

，

通常分别等于编码方法或装置中的w，x，y和z。在有损压缩的场合，以上解码方法或装置中的

，

，

，

通常或者分别等于编码方法或装置中的w，x，y，z经过变换、量化、反量化、反变换后的重构值或者经过量化、反量化后的重构值或者经过变换、反变换后的重构值。In the case of lossless compression, the above decoding method or device

,

,

,

Usually equal to w, x, y and z in the encoding method or device, respectively. In the case of lossy compression, the above decoding method or device

,

,

,

Usually or respectively equal to w, x, y, z in the encoding method or device after transformation, quantization, inverse quantization, inverse transformation reconstruction value or reconstruction value after quantization, inverse quantization or after transformation, inverse transformation the reconstructed value.

Implementation or Variation 5

In the encoding method or device or the decoding method or device described in the implementation or variant 4,

The I sets of coefficients of the linear transformation F are o[i], p[i], q[i], r[i], s[i], t[i], u[i], v[i], 0 ≤ i < I, the linear transformation F is calculated by:

y = (o[i]*w + p[i]*x + q[i])/r[i], z = (s[i]*w + t[i]*x + u[i]) /v[i];

The I set of coefficients of the linear transformation G are a[i], b[i], c[i], d[i], e[i], f[i], g[i], h[i], 0 ≤ i < I, the linear transformation G is calculated by:

= (a[i]*

+ b[i]*

+ c[i])/d[i]，

= (e[i]*

+ f[i]*

+ g[i])/h[i]。

= (a[i]*

+ b[i]*

+ c[i])/d[i],

= (e[i]*

+ f[i]*

+ g[i])/h[i].

Implementation or Variation 6

In the encoding method or device or the decoding method or device described in the implementation or variant 5, the i-th (0 ≤ i < I) set of coefficients o[i], p[i], q[i] of the forward operation of the component conversion , r[i], s[i], t[i], u[i], v[i] are abbreviated as [o, p, q, r, s, t, u, v], the Inverse operation of the ith (0 ≤ i < I) set of coefficients a[i], b[i], c[i], d[i], e[i], f[i], g[i], h [i] is abbreviated as [a, b, c, d, e, f, g, h], and the multiple sets of coefficients include at least some or all of the following seven sets of coefficients:

1) Coefficient #1: [o, p, q, r, s, t, u, v] = [1, 0, 0, 0, 0, 1, 0, 0], [a, b, c, d ,e, f, g, h] = [1, 0, 0, 0, 0, 1, 0, 0] That is, the linear transformation is an identity transformation, and the calculation formula of the forward operation using the component transformation of this set of coefficients is:

y = w, z = x;

The formula for calculating the inverse operation of component transformation is

=

，

=

；

=

,

=

;

2) Coefficient #2a: [o, p, q, r, s, t, u, v] = [1, 1, δ, 2, 1, –1, λ, 2], [a, b, c, d,e, f, g, h] = [1, 1, 0, 1, 1, –1, 0, 1], where δ=0 or 1 or –1 and λ=0 or 1 or –1 are both The rounding method controls the parameters, and the calculation formula of the forward operation using the component conversion of this set of coefficients is:

y = (w + x +δ)/2, where δ=0 or 1 or –1, z = (w - x +λ)/2, where λ=0 or 1 or –1;

The formula for calculating the inverse operation of component transformation is

=

+

，

=

-

；

=

+

,

=

-

;

3) Coefficient #2b: [o, p, q, r, s, t, u, v] = [1, –1,δ, 2, 1, 1,λ, 2], [a, b, c, d,e,f,g,h] = [1, 1, 0, 1, –1, 1, 0, 1], where δ=0 or 1 and λ=0 or 1 are rounding control parameters, use The formula for calculating the forward operation of the component conversion of this set of coefficients is:

y = (w - x + δ)/2, where δ=0 or 1 or –1, z = (w + x +λ)/2, where λ=0 or 1 or –1;

The formula for calculating the inverse operation of component transformation is

=

+

，

= -

+

；

=

+

,

= -

+

;

It can be seen that the difference between coefficient #2b and coefficient #2a is that the p of the two sets of coefficients are opposite numbers, t are opposite numbers, e are opposite numbers, and f are opposite numbers;

4) Coefficient #3a: [o, p, q, r, s, t, u, v] = [A, 2, δ, A+1, 1, –2, λ, A+1], [a, b, c, d, e, f, g, h] = [1, 1, 0, 1, 1, –A,ε, 2], where A=1 or 4 or other predetermined conditions satisfying 1≤A≤10 Integer constant, while δ, λ, ε are all rounding control parameters: δ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants that satisfy –A/2≤δ≤A/2 , λ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants satisfying –A/2≤λ≤A/2, ε=0 or 1 or –1, using this set of coefficients The formula for calculating the forward operation of the component transformation is:

y = (Aw + 2x +δ)/(A+1), z = (w - 2x +λ)/(A+1);

The formula for calculating the inverse operation of component transformation is

=

+

，

= (

- A

+ε)/2，其中ε=0或1或–1；

=

+

,

= (

-A

+ε)/2, where ε=0 or 1 or –1;

5) Coefficient #3b: [o, p, q, r, s, t, u, v] = [A, –2,δ, A+1, 1, 2,λ, A+1], [a, b, c, d, e, f, g, h] = [1, 1, 0, 1, –1, A,ε, 2], where A=1 or 4 or other predetermined conditions satisfying 1≤A≤10 Integer constant, while δ, λ, ε are all rounding control parameters: δ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants that satisfy –A/2≤δ≤A/2 , λ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants that satisfy –A/2≤λ≤A/2, ε=0 or 1 or –1, use this set of coefficients The formula for calculating the forward operation of the component transformation is:

y = (Aw - 2x +δ)/(A+1), z = (w + 2x +λ)/(A+1);

The formula for calculating the inverse operation of component transformation is

=

+

，

= (-

+ A

+ε)/2，其中ε=0或1或–1；

=

+

,

= (-

+ A

+ε)/2, where ε=0 or 1 or –1;

It can be seen that the difference between coefficient #3b and coefficient #3a is that the p of the two sets of coefficients are opposite numbers, t are opposite numbers, e are opposite numbers, and f are opposite numbers;

6) Coefficient #4a: [o, p, q, r, s, t, u, v] = [2, A, δ, A+1, –2, 1, λ, A+1], [a, b, c, d, e, f, g, h] = [1, –A,ε, 2, 1, 1, 0, 1], where A=1 or 4 or other predetermined conditions satisfying 1≤A≤10 Integer constant, while δ, λ, ε are all rounding control parameters: δ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants that satisfy –A/2≤δ≤A/2 , λ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants that satisfy –A/2≤λ≤A/2, ε=0 or 1 or –1, use this set of coefficients The formula for calculating the forward operation of the component transformation is:

y = (2w + Ax +δ)/(A+1), z = (-2w + x +λ)/(A+1);

The formula for calculating the inverse operation of component transformation is

= (

- A

+ε)/2，其中ε=0或1或–1，

=

+

；

= (

-A

+ε)/2, where ε=0 or 1 or –1,

=

+

;

It can be seen that coefficient #4a is the result of swapping o and p, swapping s and t, swapping a and e, swapping b and f, swapping c and g, and swapping d and h in coefficient #3a;

7) Coefficient #4b: [o, p, q, r, s, t, u, v] = [–2, A, δ, A+1, 2, 1, λ, A+1], [a, b,c, d, e, f, g, h] = [–1, A,ε, 2, 1, 1, 0, 1], where A=1 or 4 or other predetermined conditions satisfying 1≤A≤10 Integer constant, while δ, λ, ε are all rounding control parameters: δ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants that satisfy –A/2≤δ≤A/2 , λ=0 or 1 or –1 or A/2 or –A/2 or other predetermined integer constants satisfying –A/2≤λ≤A/2, ε=0 or 1 or –1, using this set of coefficients The formula for calculating the forward operation of the component transformation is:

y = (-2w + Ax +δ)/(A+1), z = (2w + x +λ)/(A+1);

The formula for calculating the inverse operation of component transformation is

= (-

+ A

+ε)/2，其中ε=0或1或–1，

=

+

；

= (-

+ A

+ε)/2, where ε=0 or 1 or –1,

=

+

;

It can be seen that the difference between coefficient #4b and coefficient #4a is that the o of the two sets of coefficients are mutually opposite numbers, s are mutually opposite numbers, a are mutually opposite numbers, and b are mutually opposite numbers;

It can also be seen that coefficient #4b is the result of swapping o and p, swapping s and t, swapping a and e, swapping b and f, swapping c and g, and swapping d and h in coefficient #3b.

Implementation or Variant 7 (multiple sets of coefficients are divided into K coefficient groups allowing overlapping, one compression block using one coefficient group)

In the encoding method or apparatus or decoding method or apparatus, the multiple sets of coefficients are divided into predetermined K (2≤K≤6) groups of coefficients, that is, K coefficient groups, and the kth (0≤k<K) group of coefficients That is, the k-th coefficient group has J _k sets of coefficients, and two coefficient groups are allowed to have the same set or several sets of coefficients (for example, usually each coefficient group includes the set of coefficients representing the identity transformation); one uses the component transformation A compressed block can only use coefficients in one coefficient group: there is part or all of the information needed to indicate which coefficient group a compressed block uses, called compressed block component conversion selection information, in the compressed data stream.

Implementation or Variation 8 (multiple sets of coefficients are divided into K coefficient groups allowing overlapping, one compression block using one coefficient group)

In the encoding method or device or the decoding method or device described in the implementation or variant example 7, there is also in the compressed data code stream an information that is referred to as integral compression unit component conversion selection information within the compressed block and represents an item within the compressed block. The whole compression unit uses part or all of the information required by which set of coefficients in the coefficient group used by the compression block.

Implementation or Variation 9 (Multiple sets of coefficients are divided into K coefficient groups allowing overlapping, one coefficient group is used for one compression block)

In the encoding method or device or the decoding method or device described in Implementation or Variation Examples 4, 5, and 6, the multiple sets of coefficients are divided into predetermined K (2≤K≤6) sets of coefficients, that is, K coefficient sets, and the kth set of coefficients is (0 ≤ k < K) group of coefficients, that is, the k-th coefficient group has J _k sets of coefficients, and the two coefficient groups are allowed to have the same set or several sets of coefficients (for example, usually each coefficient group includes an identity transformation. that set of coefficients); a compressed block using component transform can only use coefficients within one set of coefficients: there is a set of coefficients in the compressed data stream called compressed block component transform selection information that indicates which set of coefficients a compressed block uses some or all of the information required.

Implementation or Variation 10 (multiple sets of coefficients are divided into K coefficient groups allowing overlapping, one coefficient group is used for one compression block)

In the encoding method or device or the decoding method or device described in Implementation or Variation Example 9, the compressed data code stream also exists in the compressed data stream, which is referred to as the selected information for the conversion of integral compression unit components in the compressed block. The whole compression unit uses part or all of the information required by which set of coefficients in the coefficient group used by the compression block.

Implementation or Variation 11 (multiple sets of coefficients are divided into K coefficient groups allowing overlapping, one compression block using one coefficient group)

In the encoding method or device or the decoding method or device described in Implementation or Variation Example 6, the seven sets of coefficients are divided into predetermined K (2≤K≤6) sets of coefficients, that is, K coefficient sets, each coefficient set has 4 sets of coefficients set of coefficients, the 4 sets of coefficients include coefficient #1; a compressed block using component transform can only use coefficients within one coefficient group: there is a representation in the compressed data bitstream called compressed block component transform selected information Part or all of the information required for which coefficient group a compression block uses.

Implementation or Variation 12 (multiple sets of coefficients are divided into K coefficient groups allowing overlapping, one compression block using one coefficient group)

In the encoding method or device or the decoding method or device described in the implementation or variant 11, the seven sets of coefficients are divided into the following two sets of coefficients, that is, two coefficient groups:

Coefficient group 1: coefficient #1, coefficient #2a, coefficient #3a, coefficient #4a;

Coefficient group 2: coefficient #1, coefficient #2b, coefficient #3b, coefficient #4b.

Implementation or Variation 13 (Component Transformation and Energy-Based Quantization and Adjustment of Inverse Quantization Degree)

In the encoding method or device or decoding method or device, according to the energy of the component conversion, the quantization degree of the converted component and the corresponding inverse quantization degree are adjusted to increase the quantization and the corresponding inverse quantization degree or reduce the quantization and the corresponding inverse quantization degree. Corresponding inverse quantization degree.

Implementation or Variation 14 (Component Transformation and Energy-Based Quantization and Adjustment of Inverse Quantization Degree)

In the encoding method or device or the decoding method or device, according to the energy of the component conversion, the quantization degree of the converted component and the corresponding inverse quantization degree are adjusted, and the increased amount of the quantization and the corresponding inverse quantization degree and the energy of the component conversion are adjusted. The amount of gain is proportional to the amount of quantization and the corresponding reduction in inverse quantization is proportional to the amount of attenuation of the energy of the component conversion.

Implementation or Variation 15 (Component Transformation and Non-Energy-Based Quantization and Control of Inverse Quantization Degree)

In the encoding method or device or decoding method or device, the quantization and the corresponding inverse quantization degree are represented and controlled by the quantization parameter QP; increasing the QP plays a role in increasing the quantization and the corresponding inverse quantization degree, and reducing the QP plays the role of increasing the quantization and the corresponding inverse quantization degree. Reduce the effect of quantization and corresponding inverse quantization degree; the encoding method or device controls the quantization parameter QP based on the energy of component conversion and/or other predetermined factors including improving encoding efficiency and/or reducing encoding and decoding complexity the size of.

Implementation or Variation 16 (Component Transformation and Non-Energy-Based Quantization and Inverse Quantization Degree Adjustment by QP)

In the encoding method or device or the decoding method or device described in the implementation or variant 15, the adjustment amount of QP is represented by a QP offset or QP increment.

Implementation or variant 17 (component conversion combined with quantization and inverse quantization into normal quantization components and enhanced quantization components)

In the encoding method or device or decoding method or device, each component obtained after non-identity conversion is divided into a normal quantization component and an enhanced quantization component, and the quantization degree of the enhanced quantization component and the corresponding inverse quantization degree are compared. The quantization degree of the normal quantization component and the corresponding inverse quantization degree are several times larger than predetermined.

Implementation or variant 18 (component conversion is combined with quantized and inverse quantized QP into normal quantized components and enhanced quantized components)

In the encoding method or device or the decoding method or device described in the implementation or variant 17, the quantization and the corresponding inverse quantization degree are represented and controlled by the quantization parameter QP, and increasing the QP plays a role in increasing the quantization and the corresponding inverse quantization degree, Reducing the QP plays a role in reducing the degree of quantization and corresponding inverse quantization, and the QP value used for quantizing the enhanced quantization component and corresponding inverse quantization is higher than that used for quantizing the normal quantization component and corresponding inverse quantization. Use a QP value greater than at least 6.

Implementation or variant 19 (component conversion combined with quantized and inverse quantized QP is divided into normal quantized components and enhanced quantized components)

In the encoding method or device or the decoding method or device described in the implementation or variant examples 4, 5 and 6, the component y obtained after the non-identity transformation is a normal quantization component, and the component z is an enhanced quantization component; quantizing z and The QP value used for the corresponding inverse quantization is at least 6 larger than the QP value used to quantize y and the corresponding inverse quantization.

Implementation or Variation Example 20 (Example of Energy Calculation for Component Transformation)

In the encoding method or device or the decoding method or device described in the implementation or variant examples 5 and 6, the energy Ey of the component y obtained after linear transformation is the square root of o[i] ² + p[i] ² , the energy of the component z Ex is the square root of s[i] ² + t[i] ² .

Implementation or Variation 21 (QP offset value for QP adjustment)

In the encoding method or device or the decoding method or device described in the implementation or variant 6,

With the QP value used for quantization or inverse quantization of components not subject to component conversion or identity-converted components, the base QP _base ,

The QP value used for quantizing or inverse quantizing the component y obtained by the non-identity transformation with the i-th set of coefficients is

QPy[i] = QP _base + DQPy[i],

The QP value used for quantizing or inverse quantizing the component z obtained by the non-identity transformation with the i-th set of coefficients is

QPz[i] = QP _base + DQPz[i],

Wherein, each i-th set of coefficients has its own independent QP offset values DQPy[i] and DQPz[i].

Implementation or Variation Example 22 (Relationship between QP Offset Value and Equal Quantization Component Coefficient, Normal Enhanced Quantization Component Coefficient)

In the encoding method or device or the decoding method or device described in the implementation or variant 21,

The value of DQPy[i] is –4 or –3 or –2 or –1 or 0 or 1 or 2 or 3;

The value of DQPz[i] is DQPy[i] or DQPy[i]±1 or DQPy[i]±2 or DQPy[i]±3 or Q, where Q ≥ (DQPy[i] + 6);

The ith set of coefficients with the value of DQPz[i] as DQPy[i] or DQPy[i]±1 or DQPy[i]±2 or DQPy[i]±3 is called equal quantization with equal QP offset value component coefficients,

And the value of Q with DQPz[i], where Q ≥ (DQPy[i] + 6), the i-th set of coefficients is called the normal-enhanced quantized component coefficients with the normal-enhanced QP offset value, in this case, the components y and The components z are referred to as normal quantization components and enhanced quantization components, respectively.

Implementation or Variation Example 23 (Example of Equal Quantization Component Coefficient, Normal Enhanced Quantization Component Coefficient)

In the encoding method or device or decoding method or device described in the implementation or variant 22, several sets of coefficients in the seven sets of coefficients are combined with the QP offset value to expand into the following encoding method or device or decoding method or device. The sets of coefficients used:

1) I ₁ sets of coefficients in the seven sets of coefficients are combined with equal QP offset values to become I ₁ sets of equal quantized component coefficients,

2) The I ₂ sets of coefficients in the seven sets of coefficients are combined with the normal enhancement QP offset value to become the I ₂ sets of normal enhancement quantized component coefficients,

Thus, the encoding method or apparatus or the decoding method or apparatus uses a total of I ₁ + I ₂ sets of coefficients to perform a component transformation forward operation or an inverse operation.

Implementation or Variation 24 (Coefficient group with equal quantized component coefficients and normal enhanced quantized component coefficients)

In the encoding method or device or the decoding method or device described in the implementation or variant 23, the I ₁ + I ₂ sets of coefficients are divided into predetermined K (2≤K≤6) sets of coefficients, that is, K coefficient sets, and the kth set of coefficients is (0 ≤ k < K) group of coefficients, that is, the k-th coefficient group has J _k sets of coefficients, and the two coefficient groups are allowed to have the same set or several sets of coefficients (for example, usually each coefficient group includes an identity transformation. that set of coefficients); a compressed block using component transform can only use coefficients within one set of coefficients: the presence of information in the compressed data stream called compressed block component transform selection information indicates that a compressed block uses the K Part or all of the information required for which coefficient group among the coefficient groups.

Implementation or Variation 25 (Coefficient groups with equal quantized component coefficients and normal enhanced quantized component coefficients, each with 3 sets of coefficients)

In the encoding method or device or the decoding method or device described in the implementation or variant 23, the set of coefficients I ₁ + I ₂ is divided into predetermined K (2≤K≤6) sets of coefficients, that is, K coefficient sets, each The coefficient group has 3 sets of coefficients, the 3 sets of coefficients include coefficient #1; a compressed block using component transform can only use the coefficients in one coefficient group: there is an option called the compressed block component transform in the compressed data stream. The fixed information indicates part or all of the information required for a compression block to use which coefficient group among the K coefficient groups.

Implementation or Variation 26 (Coefficient groups with equal quantized component coefficients and normal enhanced quantized component coefficients, each with 4 sets of coefficients)

In the encoding method or device or the decoding method or device described in the implementation or variant 23, the set of coefficients I ₁ + I ₂ is divided into predetermined K (2≤K≤6) sets of coefficients, that is, K coefficient sets, each The coefficient group has 4 sets of coefficients, which include coefficient #1; a compressed block using component transform can only use coefficients in one coefficient group: there is an option called the compressed block component transform in the compressed data stream. The fixed information indicates part or all of the information required for a compression block to use which coefficient group among the K coefficient groups.

Implementation or Variation 27 (Coefficient groups with equal quantized component coefficients and normal enhanced quantized component coefficients, each with 5 sets of coefficients)

In the encoding method or device or the decoding method or device described in the implementation or variant 23, the set of coefficients I ₁ + I ₂ is divided into predetermined K (2≤K≤6) sets of coefficients, that is, K coefficient sets, each The coefficient group has 5 sets of coefficients, which include coefficient #1; a compressed block using component transform can only use coefficients in one coefficient group: there is an option called the compressed block component transform in the compressed data stream. The fixed information indicates part or all of the information required for a compression block to use which coefficient group among the K coefficient groups.

Implementation or Variation 28 (Coefficient groups with equal quantized component coefficients and normal enhanced quantized component coefficients, each with 3 sets of coefficients)

In the encoding method or device or the decoding method or device described in the implementation or variant 24, 25, 26, and 27, there is also a representation of the selected information called integral compression unit component conversion within the compressed block in the compressed data stream. An integral compression unit within a compression block uses some or all of the information required by which set of coefficients within that coefficient set used by the compression block.

Implementation or Variation Example 29 (Example of Coefficient Group Containing Equal Quantization Component Coefficients and Normal Enhanced Quantization Component Coefficients)

In the encoding method or apparatus or the decoding method or apparatus described in the implementation or variant 24, the K coefficient groups are K coefficient groups selected from the following coefficient groups:

Coefficient group 1: coefficient #1, coefficient #2a combined with equal QP offset value, coefficient #3a combined with equal QP offset value, coefficient #4a combined with equal QP offset value;

Coefficient group 2: coefficient #1, coefficient #2b combined with equal QP offset value, coefficient #3b combined with equal QP offset value, coefficient #4b combined with equal QP offset value;

Coefficient group 3: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #3a combined with normal enhancement QP offset value, coefficient #4a combined with normal enhancement QP offset value;

Coefficient group 4: coefficient #1, coefficient #2b combined with normal enhancement QP offset value, coefficient #3b combined with normal enhancement QP offset value, coefficient #4b combined with normal enhancement QP offset value;

Coefficient group 5: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #2a combined with equal QP offset value, coefficient #3a combined with normal enhancement QP offset value;

Coefficient Group 6: Coefficient #1, Coefficient #2a combined with Normal Augmented QP Offset, Coefficient #2a coupled with Equal QP Offset, Coefficient #4a coupled with Normal Augmented QP Offset;

Coefficient group 7: coefficient #1, coefficient #2b combined with normal enhanced QP offset value, coefficient #2b combined with equal QP offset value, coefficient #3b combined with normal enhanced QP offset value;

Coefficient group 8: coefficient #1, coefficient #2b combined with normal enhanced QP offset value, coefficient #2b combined with equal QP offset value, coefficient #4b combined with normal enhanced QP offset value;

Coefficient Set 9: Coefficient #1, Coefficient #2a combined with Normal Enhancement QP Offset, Coefficient #2a combined with Equal QP Offset, Coefficient #3a combined with Normal Enhanced QP Offset, and Normal Enhanced QP Coefficient #4a of the offset value combination;

Coefficient set 10: Coefficient #1, Coefficient #2b combined with Normal Enhancement QP Offset, Coefficient #2b combined with Equal QP Offset, Coefficient #3b combined with Normal Enhanced QP Offset, and Normal Enhanced QP Coefficient #4b of offset value combination;

Coefficient Set 11: Coefficient #1, Coefficient #2a combined with Normal Enhanced QP Offset, Coefficient #2a combined with Equal QP Offset, Coefficient #3a combined with Normal Enhanced QP Offset, and Equal QP Offset Coefficient #3a combined with shift value, Coefficient #4a combined with normal enhanced QP offset value, Coefficient #4a combined with equal QP offset value;

Coefficient set 12: Coefficient #1, Coefficient #2b combined with Normal Enhanced QP Offset, Coefficient #2b combined with Equal QP Offset, Coefficient #3b combined with Normal Enhanced QP Offset, and Equal QP Offset Coefficient #3b combined with shift value, coefficient #4b combined with normal enhanced QP offset value, coefficient #4b combined with equal QP offset value.

Implementation or Variation Example 30 (Example of a coefficient group including equal quantization component coefficients and normal enhanced quantization component coefficients, each coefficient group has 4 sets of coefficients)

In the encoding method or device or the decoding method or device described in implementation or variant 26, the K coefficient groups are K coefficient groups selected from the following coefficient groups:

Coefficient group 1: coefficient #1, coefficient #2a combined with equal QP offset value, coefficient #3a combined with equal QP offset value, coefficient #4a combined with equal QP offset value;

Coefficient group 2: coefficient #1, coefficient #2b combined with equal QP offset value, coefficient #3b combined with equal QP offset value, coefficient #4b combined with equal QP offset value;

Coefficient group 3: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #3a combined with normal enhancement QP offset value, coefficient #4a combined with normal enhancement QP offset value;

Coefficient group 4: coefficient #1, coefficient #2b combined with normal enhancement QP offset value, coefficient #3b combined with normal enhancement QP offset value, coefficient #4b combined with normal enhancement QP offset value;

Coefficient group 5: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #2a combined with equal QP offset value, coefficient #3a combined with normal enhancement QP offset value;

Coefficient Group 6: Coefficient #1, Coefficient #2a combined with Normal Augmented QP Offset, Coefficient #2a coupled with Equal QP Offset, Coefficient #4a coupled with Normal Augmented QP Offset;

Coefficient group 7: coefficient #1, coefficient #2b combined with normal enhanced QP offset value, coefficient #2b combined with equal QP offset value, coefficient #3b combined with normal enhanced QP offset value;

Coefficient set 8: Coefficient #1, Coefficient #2b combined with normal enhancement QP offset value, Coefficient #2b combined with Equal QP offset value, Coefficient #4b combined with normal enhancement QP offset value.

Implementation or Variation Example 31 (Example of a coefficient group including equal quantization component coefficients and normal enhancement component coefficients, each coefficient group has 4 sets of coefficients)

In the encoding method or apparatus or the decoding method or apparatus described in the implementation or variant 26, the K coefficient groups are the following 2 coefficient groups:

Coefficient group A: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #3a combined with normal enhancement QP offset value, coefficient #4a combined with normal enhancement QP offset value;

Coefficient set B: coefficient #1, coefficient #2b combined with normal enhancement QP offset value, coefficient #3b combined with normal enhancement QP offset value, coefficient #4b combined with normal enhancement QP offset value.

Implementation or Variation Example 32 (Example of a coefficient group including equal quantization component coefficients and normal enhanced quantization component coefficients, each coefficient group has 4 sets of coefficients)

In the encoding method or apparatus or the decoding method or apparatus described in the implementation or variant 26, the K coefficient groups are the following 4 coefficient groups:

Coefficient Group I: Coefficient #1, Coefficient #2a combined with Normal Augmented QP Offset, Coefficient #2a coupled with Equal QP Offset, Coefficient #3a coupled with Normal Augmented QP Offset;

Coefficient group II: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #2a combined with equal QP offset value, coefficient #4a combined with normal enhancement QP offset value;

Coefficient group III: coefficient #1, coefficient #2b combined with normal enhancement QP offset value, coefficient #2b combined with equal QP offset value, coefficient #3b combined with normal enhancement QP offset value;

Coefficient set IV: Coefficient #1, Coefficient #2b combined with normal enhancement QP offset value, Coefficient #2b combined with Equal QP offset value, Coefficient #4b combined with normal enhancement QP offset value.

Implementation or Variation Example 33 (The QP of quantization and inverse quantization of the enhanced quantization component is large, so the enhanced quantization component after quantization and inverse quantization is zero)

的值都是零，仅从正常量化分量

计算获得分量

和分量

。Implement or variant 4, 5, 6, 8, 10, 11, 12, 16, 18, 21, 22, 23, 24, 25, 26, 27, 29, 30, 31, 32 The encoding method or apparatus described in or In the decoding method or device, the quantization and the corresponding inverse quantization degree are represented and controlled by the quantization parameter QP; increasing the QP can increase the quantization and the corresponding inverse quantization degree, and reducing the QP can reduce the quantization and the corresponding inverse quantization degree. The function of quantization degree; the components obtained after non-identity transformation are divided into normal quantization components and enhanced quantization components, and the QP values used for quantizing the enhanced quantization components and corresponding inverse quantization are compared with the normal quantization components. The QP value used for the quantization of the components and the corresponding inverse quantization is at least 20, so the enhanced quantization components are all zero after quantization, and the reconstructed values of the residual data obtained after at least inverse quantization are also all zero, Therefore, in the inverse operation of the component transform, the quantized components are strengthened

The values are all zero, only from the normal quantization component

Calculate the weight

and weight

.

Implementation or Variation Example 34 (The QP of quantization and inverse quantization of the enhanced quantization component is large, so the enhanced quantization component after quantization and inverse quantization is zero)

进行反量化所使用的QP值比对所述正常量化分量

进行反量化所使用的QP值大至少20，因而对应的强化量化分量z在对应的编码方法或装置中经过量化后全部为零，因此，所述强化量化分量

的值都是零，在分量变换的逆操作中，仅从正常量化分量

计算获得分量

和分量

。In the implementation or variant 6, 11, 12, 21, 22, 23, 24, 25, 26, 27, 29, 30, 31, 32 of the decoding method or device, the enhanced quantization component is

The QP value used for inverse quantization is compared with the normal quantization component

The QP value used for inverse quantization is at least 20, so the corresponding enhanced quantization component z is all zero after being quantized in the corresponding encoding method or device. Therefore, the enhanced quantized component z is

The values of are all zero, and in the inverse operation of the component transform, only the normal quantized components are

Calculate the weight

and weight

.

Implementation or Variation Example 35 (The QP of quantization and inverse quantization of the enhanced quantization component is large, so the enhanced quantization component after quantization and inverse quantization is zero)

In the decoding method or device described in the implementation or variant 34, the calculation formulas for the inverse operation of the component transformation in the seven sets of coefficients are respectively:

1) Coefficient #1: still identity transformation;

=

，

=

；2) Coefficient #2a:

=

,

=

;

=

，

= -

；3) Coefficient #2b:

=

,

= -

;

=

，

= (

+ε)/2，其中ε=0或1或–1；4) Coefficient #3a:

=

,

= (

+ε)/2, where ε=0 or 1 or –1;

=

，

= (-

+ε)/2，其中ε=0或1或–1；5) Coefficient #3b:

=

,

= (-

+ε)/2, where ε=0 or 1 or –1;

= (

+ε)/2，其中ε=0或1或–1，

=

；6) Coefficient #4a:

= (

+ε)/2, where ε=0 or 1 or –1,

=

;

= (-

+ε)/2，其中ε=0或1或–1，

=

。7) Coefficient #4b:

= (-

+ε)/2, where ε=0 or 1 or –1,

=

.

Implementation or Variation 36 (Implementation or Variation on Selected Information)

In the encoding method or device or the decoding method or device, the selected information exists in the compressed block header and/or the entire compressed unit header in a direct form, an indirect form, or a mixed form of direct and indirect; The information consists of one or more bit strings (bit strings) in the compressed data code stream, the selected information in the indirect form is derived from other encoding parameters and/or codec variables and/or other syntax of the compressed data code stream element-derived information, the selected information of the direct-indirect mix is partially direct (that is, consists of one or more bit strings in the compressed data codestream) and partially indirect (that is, from other encoding parameters and/or codec variables and and/or other syntax elements derived from the compressed data codestream) selected information for mixing.

Implementation or Variation 37 (Implementation or Variation on Selected Information)

In the encoding method or device or the decoding method or device of the implementation or variant 36, the compressed block header is a sequence parameter set or a picture parameter set or a sequence header or a picture header or a slice header or a slice header or a brick header or a maximum coding unit LCU header or Coding Tree Unit CTU header or Coding Unit CU header.

Implementation or Variation 38 (Implementation or Variation on Selected Information)

In the encoding method or device or the decoding method or device, one or more flag bits and/or one or more identification codes, which are directly or indirectly or directly mixed in italics, are obtained from at least the selected information: and associated syntax elements:

component conversion selected coefficient flag bit and/or component conversion selected coefficient identification code;

The direct flag bits and/or identification codes are composed of one or more bit strings (bit strings) in the compressed data stream, and the indirect flag bits and/or identification codes are obtained from other encoding parameters and/or Codec variables and/or flags and/or identification codes derived from other syntax elements of the compressed data codestream, the directly and indirectly mixed or multiple bit strings) partially indirectly (ie derived from other encoding parameters and/or codec variables and/or other syntax elements of the compressed data stream) mixed flag bits and/or identification codes.

Implementation or Variation 39 (Implementation or Variation on Selected Information)

In the encoding method or device or the decoding method or device described in the implementation or variant 38, the multiple sets of coefficients are I (I ≥ 3) sets of coefficients, which are respectively referred to as the i-th (0 ≤ i < I) set of coefficients; The component conversion and/or component conversion selected coefficient identification codes take predetermined II (II ≥ I) values, which are respectively referred to as the iith (0 ≤ ii < II) value; each of the iith values corresponds to A set of coefficients predetermined in the I set of coefficients is called the i(ii) set of coefficients, and different ii are allowed to correspond to the same i(ii), such as: i(0)=i(1)=i (2)=i(3)=0, that is, the 0th, 1st, 2nd, and 3rd values of the component conversion selected coefficient identification code all correspond to the 0th set of coefficients; at least according to the component conversion Selected coefficient flags and/or component transforms the value of the selected coefficient identification code, performing the following corresponding component transforms:

If the value of the component transformation selected coefficient flag bit and/or the component transformation selected coefficient identification code is equal to the iith value, then {use the i(ii)th set of coefficients to perform a component transformation positive operation on the residual Or the inverse of component transformation}.

Implementation or Variation 40 (Implementation or Variation on Compressed Block Component Conversion Selected Information)

In the encoding method or device or the decoding method or device described in the implementation or variant 11, 24, 25, 26, 27, the compressed block component conversion selected information exists in the compressed form in direct form or indirect form or direct-indirect mixed form In the block header; the selected information in the direct form consists of one or more bit strings (bit strings) in the compressed data stream, and the selected information in the indirect form is derived from other encoding parameters and/or codecs Information derived from variables and/or other syntax elements of the compressed data codestream, the selected information for the direct-indirect mix is part direct (i.e. consisting of one or more bit strings in the compressed data codestream) part indirect (i.e. derived from other coding parameters and/or codec variables and/or other syntax elements of the compressed data codestream) mixed selected information.

Implementation or Variation 41 (Implementation or Variation on Compressed Block Component Conversion Selection Information)

In the encoding method or device or the decoding method or device of the implementation or variant 40, the compressed block header is a sequence parameter set or a picture parameter set or a sequence header or a picture header or a slice header or a slice header or a brick header or a maximum coding unit LCU header or Coding Tree Unit CTU header or Coding Unit CU header.

Implementation or Variation 42 (Implementation or Variation on Compressed Block Component Conversion Selection Information)

In implementations or variants 11, 24, 25, 26, 27 of the encoding method or device or decoding method or device, the following direct or indirect or direct representations in italics are obtained from at least the compressed block component conversion selection information: One or more flag bits and/or one or more identifiers and associated syntax elements mixed in indirectly:

compressed block component conversion selected coefficient flag bit and/or compressed block component conversion selected coefficient identification code;

The direct flag bits and/or identification codes are composed of one or more bit strings (bit strings) in the compressed data stream, and the indirect flag bits and/or identification codes are obtained from other encoding parameters and/or Codec variables and/or flags and/or identification codes derived from other syntax elements of the compressed data codestream, the directly and indirectly mixed or multiple bit strings) partially indirectly (ie derived from other encoding parameters and/or codec variables and/or other syntax elements of the compressed data stream) mixed flag bits and/or identification codes.

Implementation or Variation 43 (Implementation or Variation on Compressed Block Component Conversion Selection Information)

In the encoding method or device or the decoding method or device described in Implementation or Variation 42, the K coefficient groups are respectively referred to as the kth (1 ≤ k ≤ K) coefficient group; the compressed block component conversion selects The coefficient flag bit and/or the coefficient identification code selected for the conversion of the compressed block components takes predetermined K values, which are respectively referred to as the kth (0 ≤ k < K) value; at least according to the compressed block component conversion and selected The coefficient flags and/or the compressed block component transforms the value of the selected coefficient identification code, performing the following corresponding component transforms:

If the value of the compressed block component transform selected coefficient flag and/or the identification code is equal to the k th value, then

{Use the coefficients belonging to the k-th coefficient group to perform a forward component transformation operation or an inverse component transformation operation on the residuals}.

Implementation or Variation 44 (Implementation or Variation of Selected Information on Integral Compression Unit Component Conversion within a Compressed Block)

In the encoding method or device or the decoding method or device described in the implementation or variant 43, there is also information in the compressed data code stream that represents the compressed block (that is, using An integral compression unit in the compression block of the k-th coefficient group) uses part or all of the information required for which set of coefficients in that coefficient group used by the compression block, from at least the compression block Intra-integration of compression unit components transforms the selected information to obtain one or more flag bits and/or one or more identification codes and associated syntax elements, directly or indirectly or directly mixed, in italics:

Integral Compression Unit Component Transform Selected Coefficient Flags in Compression Block

and / or

Integer Compression Unit Component Conversion Selected Coefficient Identification Code in the Compression Block

The direct flag bits and/or identification codes are composed of one or more bit strings (bit strings) in the compressed data stream, and the indirect flag bits and/or identification codes are obtained from other encoding parameters and/or Codec variables and/or flags and/or identification codes derived from other syntax elements of the compressed data codestream, the directly and indirectly mixed or multiple bit strings) partially indirectly (ie derived from other encoding parameters and/or codec variables and/or other syntax elements of the compressed data stream) mixed flag bits and/or identification codes.

Implementation or Variation 45 (Implementation or Variation of Selected Information on Integral Compression Unit Component Conversion within a Compressed Block)

In the encoding method or device or the decoding method or device described in the implementation or variant 44, the k th coefficient group has J _k sets of coefficients, which are respectively referred to as the j th (0 ≤ j < J _k ) set of coefficients; the The flag bits of the coefficients selected for the conversion of the integral compression unit components in the compression block and/or the identification codes of the coefficients selected for the conversion of the integral compression unit components in the compression block take predetermined values of JJ _k (JJ _k ≥ J _k ), which are respectively referred to as The jjth (0 ≤ jj < JJ _k ) value; each of the jjth values corresponds to a predetermined set of coefficients in the _Jk set of coefficients, which is called the j(jj)th set of coefficients, different jj It is allowed to correspond to the same j(jj), such as: j(0)=j(1)=j(2)=j(3)=0, that is, the selected coefficient identifier of the integral compression unit component conversion in the compression block The four values of the 0th, 1st, 2nd, and 3rd values of the code all correspond to the 0th set of coefficients; at least according to the integral compression unit component conversion in the compression block, the selected coefficient flag bit and/or the integral in the compression block The compression unit component transforms the value of the selected coefficient identification code, performing the following corresponding component transforms:

If the value of the selected coefficient flag bit and/or the identification code of the integral compression unit component transform in the compressed block is equal to the jjth value, then {use the j(jj)th set of coefficients to component transform the residual Forward operation or inverse operation of component transformation}.

Implementation or Variation Example 46 (Example of Integer Compression Unit Component Conversion Selected Information in Compression Block, K=2, _Jk= 4)

In the encoding method or apparatus or the decoding method or apparatus described in the implementation or variant 45, the K coefficient groups are the following 2 coefficient groups:

Coefficient group A: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #3a combined with normal enhancement QP offset value, coefficient #4a combined with normal enhancement QP offset value;

Coefficient set B: coefficient #1, coefficient #2b combined with normal enhancement QP offset value, coefficient #3b combined with normal enhancement QP offset value, coefficient #4b combined with normal enhancement QP offset value.

Implementation or Variation Example 47 (Example of Integer Compression Unit Component Conversion Selected Information in Compression Block, K=2, _Jk= 4)

In the encoding method or device or the decoding method or device described in the implementation or variant 46, the compressed block component conversion selection information is the compressed block component conversion selected coefficient flag or the sign flag (name from The difference between coefficient #2a, coefficient #3a, coefficient #4a and coefficient #2b, coefficient #3b, coefficient #4b is that several coefficients differ by a sign, that is, they are opposite numbers); The fixed coefficient flag or the sign flag exists in the compressed block header in a direct form or an indirect form or a direct-indirect mixed form; the compressed block component in the direct form converts the selected coefficient flag or the sign flag The bits consist of one or more bit strings (bit strings) in the compressed data stream, and the indirect form of the compressed block component conversion selected coefficient flag or sign flag is derived from other encoding parameters and/or Compressed block components derived from codec variables and/or other syntax elements of the compressed data codestream convert the selected coefficient flag or sign flag, and the directly-indirectly mixed compressed block component converts the selected coefficient flag or The sign flag is partially direct (that is, consisting of one or more bit strings in the compressed data stream) and partially indirect (that is, derived from other encoding parameters and/or codec variables and/or other syntax of the compressed data stream Element export) mixed compressed block component transforms the selected coefficient flag or the sign flag.

Implementation or Variation Example 48 (Example of Integral Compression Unit Component Conversion Selected Information in Compression Block, K=2, _Jk= 4)

In the encoding method or device or decoding method or device described in the implementation or variant 47, the compressed block component conversion selected coefficient flag or the sign flag takes two predetermined values: a first predetermined value and a first predetermined value. Two predetermined values; at least according to the value of the selected coefficient flag bit or the sign flag bit of the compressed block component conversion, perform the following corresponding component conversion:

If the compressed block component-transform selected coefficient flag or the sign flag is equal to the first predetermined value, then {use the coefficients belonging to the coefficient group A to perform a component-transform positive operation or component on the residual Convert inverse operation}

If the compressed block component transform selected coefficient flag or the sign flag is equal to the second predetermined value, then {use the coefficients belonging to the coefficient group B to perform a component transform positive operation or component on the residual Convert inverse operation}.

Implementation or Variation Example 49 (Example of Integer Compression Unit Component Conversion Selected Information in Compression Block, K=2, _Jk= 4)

In the encoding method or device or the decoding method or device described in the embodiment or variant 48, the selected information for the conversion of the integral compression unit component in the compression block is the selected coefficient identification code for the conversion of the integral compression unit component in the compression block or The chroma residual coding and component conversion type; the integral compression unit component conversion selected coefficient identification code in the compression block or the chroma residual coding and component conversion type exists in direct form or indirect form or direct indirect mixed form In the integral compression unit header; the integral compression unit component conversion selected coefficient identification code or chroma residual coding and component conversion type in the compression block of the direct form is determined by one or more bit strings in the compressed data code stream ( bit string), the indirect form of the compression block within the integral compression unit component conversion selected coefficient identification code or chroma residual coding and the component conversion type is from other coding parameters and/or codec variables and/or compression Intra-compressed unit component conversion in compressed blocks derived from other syntax elements of the data stream selects the coefficient identification code or chroma residual coding and component conversion type, and the directly-indirectly mixed intra-compressed intra-compression unit component conversion selects Fixed coefficient identification codes or chroma residual coding and component conversion types are partially direct (that is, composed of one or more bit strings in the compressed data stream) and partially indirect (that is, derived from other encoding parameters and/or codec variables and / or other syntax elements derived from the compressed data codestream) mixed compressed intra-block integral compression unit component transform selected coefficient identification code or chroma residual coding and component transform type.

Implementation or Variation Example 50 (Example of Integral Compression Unit Component Conversion Selected Information in Compression Block, K=2, _Jk= 4)

In the encoding method or device or the decoding method or device described in Implementation or Variation 49, the selected coefficient identification code for integral compression unit component conversion in the compression block or the chrominance residual coding and component conversion type are selected from seven predetermined values. Value: Predetermined value 1, Predetermined value 2, Predetermined value 3, Predetermined value 4, Predetermined value 5, Predetermined value 6, Predetermined value 7; The values of the chroma residual coding and component transform types described above, perform the following corresponding component transforms:

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 1, then

和分量

都为零，因此，实际上不需要对分量

和分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

and weight

are all zero, so there is really no need for

and weight

do residual decoding}

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 2, then

为零而分量

不为零，因此，实际上不需要对分量

进行残差解码而仅需要对分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

zero and component

is not zero, so there is actually no need for

perform residual decoding and only need to

do residual decoding}

If the Intra Compression Unit Component Transform Selected Coefficient Identification Code or the Chroma Residual Coding and Component Transform Type is equal to the predetermined value of 3, then

不为零而分量

为零，因此，实际上仅需要对分量

进行残差解码而不需要对分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

non-zero but component

is zero, so in practice only the component

perform residual decoding without

do residual decoding}

If the Intra Compression Unit Component Transform Selected Coefficient Identification Code or the Chroma Residual Coding and Component Transform Type is equal to the predetermined value of 4, then

和分量

都不为零，因此，实际上需要对分量

和分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

and weight

are not zero, so it is actually necessary to

and weight

do residual decoding}

If the Intra Compression Unit Component Transform Selected Coefficient Identification Code or the Chroma Residual Coding and Component Transform Type is equal to the predetermined value of 5, then

{

If the compressed block component transform selected coefficient flag or the sign flag is equal to the first predetermined value, then {component transform the residual using coefficient #2a combined with the normal enhanced QP offset value Forward or Component Transform Inverse }; else {Component Transform Forward or Component Transform Inverse on Residual Using Coefficient #2b Combined with Normal Augmented QP Offset Value}

}

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 6, then

{

If the compressed block component transform selected coefficient flag or the sign flag is equal to the first predetermined value, then {component transform the residual using coefficient #3a combined with the normal enhanced QP offset value Forward or Component Transform Inverse }; else {Component Transform Forward or Component Transform Inverse on Residual Using Coefficient #3b Combined with Normal Augmented QP Offset Value}

}

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 7, then

{

If the compressed block component transform selected coefficient flag or the sign flag is equal to the first predetermined value, then {component transform the residual using coefficient #4a combined with the normal enhanced QP offset value Forward or Component Transform Inverse }; else {Component Transform Forward or Component Transform Inverse on Residual Using Coefficient #4b Combined with Normal Augmented QP Offset Value}

}.

Implementation or Variation Example 51 (Example of Integer Compression Unit Component Conversion Selection Information in Compression Block, K=4, _Jk =4)

In the encoding method or apparatus or the decoding method or apparatus described in the implementation or variant 45, the K coefficient groups are the following 4 coefficient groups:

Coefficient Group I: Coefficient #1, Coefficient #2a combined with Normal Augmented QP Offset, Coefficient #2a coupled with Equal QP Offset, Coefficient #3a coupled with Normal Augmented QP Offset;

Coefficient group II: coefficient #1, coefficient #2a combined with normal enhancement QP offset value, coefficient #2a combined with equal QP offset value, coefficient #4a combined with normal enhancement QP offset value;

Coefficient group III: coefficient #1, coefficient #2b combined with normal enhancement QP offset value, coefficient #2b combined with equal QP offset value, coefficient #3b combined with normal enhancement QP offset value;

Coefficient set IV: Coefficient #1, Coefficient #2b combined with normal enhancement QP offset value, Coefficient #2b combined with Equal QP offset value, Coefficient #4b combined with normal enhancement QP offset value.

Implementation or Variation Example 52 (Example of Integer Compression Unit Component Conversion Selection Information in Compression Block, K=4, _Jk =4)

In the encoding method or device or the decoding method or device described in the implementation or variant 51, the compressed block component conversion selection information is the compressed block component conversion selected coefficient identification code or sign flag (name from The difference between coefficient #2a, coefficient #3a, coefficient #4a and coefficient #2b, coefficient #3b, coefficient #4b is that some coefficients differ by a sign, that is, they are opposite to each other) and the swap flag (the name comes from the coefficient The difference between #3a and coefficient #3b and coefficient #4a and coefficient #4b lies in swapping o and p, swapping s and t, swapping a and e, swapping b and f, swapping c and g, swapping d and h); The compressed block component conversion selected coefficient identification code or the sign flag and the swap flag exist in the compressed block header in direct form or indirect form or direct-indirect mixed form; the direct form of the compressed block component The conversion selected coefficient identification code or the sign flag and the swap flag are composed of one or more bit strings (bit strings) in the compressed data code stream, and the compression block component in the indirect form converts the selected coefficient identification The code or sign flag and the swap flag are the compressed block component transform selected coefficient flag or sign flag derived from other encoding parameters and/or codec variables and/or other syntax elements of the compressed data codestream bit and the swap flag, the directly-indirectly mixed compressed block component conversion selected coefficient identification code or sign flag and the swap flag are partially direct (i.e. by one or more bit strings in the compressed data stream Compressed block components that are mixed in part indirectly (i.e. derived from other coding parameters and/or codec variables and/or other syntax elements of the compressed data codestream) convert the selected coefficient identification code or sign flag and swap flag bit.

Implementation or Variation Example 53 (Example of Integral Compression Unit Component Conversion Selected Information in Compression Block, K=4, _Jk =4)

In the encoding method or device or the decoding method or device described in the implementation or variant 52, the compression block component conversion selected coefficient identification code takes four predetermined values: predetermined value one, predetermined value two, predetermined value three, predetermined value Four, or the sign flag bit and the swap flag bit take two predetermined values independently of each other: a first predetermined value and a second predetermined value; at least according to the compression block component conversion selected coefficient identification code or all According to the value of the sign flag and the swap flag, the following corresponding component conversions are performed:

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value one or the sign flag is equal to the first predetermined value and the swap flag is also equal to the first predetermined value, then

{Use the coefficients belonging to the coefficient group I to perform a component transformation forward operation or a component transformation inverse operation on the residuals}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value of two or the sign flag is equal to the first predetermined value and the swap flag is equal to the second predetermined value, then

{Use the coefficients belonging to the coefficient group II to perform a forward component transformation operation or an inverse component transformation operation on the residuals}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value three or the sign flag is equal to the second predetermined value and the swap flag is equal to the first predetermined value, then

{Use the coefficients belonging to the coefficient group III to perform a forward component transformation operation or an inverse component transformation operation on the residuals}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value four or the sign flag is equal to the second predetermined value and the swap flag is also equal to the second predetermined value, then

{Use the coefficients belonging to the coefficient group IV to perform a forward component transformation operation or an inverse component transformation operation on the residuals}.

Implementation or Variation Example 54 (Example of Integral Compression Unit Component Conversion Selected Information in Compression Block, K=4, _Jk =4)

In the encoding method or device or the decoding method or device described in the implementation or variant 53, the selected information for the conversion of the integral compression unit component in the compression block is the selected coefficient identification code for the conversion of the integral compression unit component in the compression block or The chroma residual coding and component conversion type; the integral compression unit component conversion selected coefficient identification code in the compression block or the chroma residual coding and component conversion type exists in direct form or indirect form or direct indirect mixed form In the integral compression unit header; the integral compression unit component conversion selected coefficient identification code or chroma residual coding and component conversion type in the compression block of the direct form is determined by one or more bit strings in the compressed data code stream ( bit string), the indirect form of the compression block within the integral compression unit component conversion selected coefficient identification code or chroma residual coding and the component conversion type is from other coding parameters and/or codec variables and/or compression Intra-compressed unit component conversion in compressed blocks derived from other syntax elements of the data stream selects the coefficient identification code or chroma residual coding and component conversion type, and the directly-indirectly mixed intra-compressed intra-compression unit component conversion selects Fixed coefficient identification codes or chroma residual coding and component conversion types are partially direct (that is, composed of one or more bit strings in the compressed data stream) and partially indirect (that is, derived from other encoding parameters and/or codec variables and / or other syntax elements derived from the compressed data codestream) mixed compressed intra-block integral compression unit component transform selected coefficient identification code or chroma residual coding and component transform type.

Implementation or Variation Example 55 (Example of Integer Compression Unit Component Conversion Selected Information in Compression Block, K=4, _Jk =4)

In the encoding method or device or the decoding method or device described in the implementation or variant 54, the selected coefficient identification code for the component conversion of the integral compression unit in the compression block or the chroma residual coding and component conversion type are selected from seven predetermined values. Value: Predetermined value 1, Predetermined value 2, Predetermined value 3, Predetermined value 4, Predetermined value 5, Predetermined value 6, Predetermined value 7; The values of the chroma residual coding and component transform types described above, perform the following corresponding component transforms:

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 1, then

和分量

都为零，因此，实际上不需要对分量

和分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

and weight

are all zero, so there is really no need for

and weight

do residual decoding}

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 2, then

为零而分量

不为零，因此，实际上不需要对分量

进行残差解码而仅需要对分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

zero and component

is not zero, so there is actually no need for

perform residual decoding and only need to

do residual decoding}

If the Intra Compression Unit Component Transform Selected Coefficient Identification Code or the Chroma Residual Coding and Component Transform Type is equal to the predetermined value of 3, then

不为零而分量

为零，因此，实际上仅需要对分量

进行残差解码而不需要对分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

not zero but component

is zero, so in practice only the component

Perform residual decoding without the need for component

do residual decoding}

If the Intra Compression Unit Component Transform Selected Coefficient Identification Code or the Chroma Residual Coding and Component Transform Type is equal to the predetermined value of 4, then

和分量

都不为零，因此，实际上需要对分量

和分量

进行残差解码}{Use coefficient #1 to perform component transformation forward operation or component transformation inverse operation on residual, component

and weight

are not zero, so it is actually necessary to

and weight

do residual decoding}

If the Intra Compression Unit Component Transform Selected Coefficient Identification Code or the Chroma Residual Coding and Component Transform Type is equal to the predetermined value of 5, then

{

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value one or the sign flag is equal to the first predetermined value and the swap flag is also equal to the first predetermined value, then {Use the coefficient #2a combined with the normal enhancement QP offset value to perform a component transformation forward operation or a component transformation inverse operation on the residual}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value of two or the sign flag is equal to the first predetermined value and the swap flag is equal to the second predetermined value, then { Component transform forward operation or component transform inverse operation on residuals using coefficient #2a combined with normal enhancement QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value three or the sign flag is equal to the second predetermined value and the swap flag is equal to the first predetermined value, then { Component transform forward operation or component transform inverse operation on residual using coefficient #2b combined with normal enhancement QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value four or the sign flag is equal to the second predetermined value and the swap flag is also equal to the second predetermined value, then {Use the coefficient #2b combined with the normal enhancement QP offset value to perform a component transformation forward operation or a component transformation inverse operation on the residual}

}

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 6, then

{

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value one or the sign flag is equal to the first predetermined value and the swap flag is also equal to the first predetermined value, then {component transform forward operation or component transform inverse operation on residual using coefficient #2a combined with equal QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value of two or the sign flag is equal to the first predetermined value and the swap flag is equal to the second predetermined value, then { Component transform forward operation or component transform inverse operation on residual using coefficient #2a combined with equal QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value three or the sign flag is equal to the second predetermined value and the swap flag is equal to the first predetermined value, then { Component transform forward operation or component transform inverse operation on residual using coefficient #2b combined with equal QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value four or the sign flag is equal to the second predetermined value and the swap flag is also equal to the second predetermined value, then {component transform forward operation or component transform inverse operation on residual using coefficient #2b combined with equal QP offset value}

}

If the integer compression unit component transform selected coefficient identification code within the compressed block or the chroma residual coding and component transform type is equal to the predetermined value of 7, then

{

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value one or the sign flag is equal to the first predetermined value and the swap flag is also equal to the first predetermined value, then {Use the coefficient #3a combined with the normal enhancement QP offset value to perform a component transformation forward operation or a component transformation inverse operation on the residual}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value of two or the sign flag is equal to the first predetermined value and the swap flag is equal to the second predetermined value, then { Component transform forward operation or component transform inverse operation on residuals using coefficient #4a combined with normal enhancement QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value three or the sign flag is equal to the second predetermined value and the swap flag is equal to the first predetermined value, then { Component transform forward operation or component transform inverse operation on residual using coefficient #3b combined with normal enhancement QP offset value}

If the compressed block component conversion selected coefficient identification code is equal to the predetermined value four or the sign flag is equal to the second predetermined value and the swap flag is also equal to the second predetermined value, then {Use the coefficient #4b combined with the normal enhancement QP offset value to perform a component transformation forward operation or a component transformation inverse operation on the residual}

}.

Implementation or Variation Example 56 (Example of First Predetermined Value and Second Predetermined Value)

In the implementation or variant 48 or 50 or 53 or 55 of the encoding method or device or the decoding method or device,

the first predetermined value is 0, the second predetermined value is 1,

or

The first predetermined value is 1 and the second predetermined value is 0.

Implementation or Variation Example 57 (Examples of Predetermined Values 1 to 7)

In the implementation or variant 50 or 55, the encoding method or device or the decoding method or device,

The predetermined value 1 is 00,

The predetermined value 2 is 010,

The predetermined value 3 is 100,

The predetermined value 4 is 110,

The predetermined value 5 is 011,

The predetermined value 6 is 101,

the predetermined value 7 is 111;

or

The predetermined value 1 is 00,

The predetermined value 2 is 010,

The predetermined value 3 is 011,

The predetermined value 4 is 100,

The predetermined value 5 is 101,

The predetermined value 6 is 110,

The predetermined value 7 is 111.

Implementation or Variation Example 58 (Examples of Predetermined Value One, Predetermined Value Two, Predetermined Value Three, Predetermined Value Four)

In the implementation or variant 53 or 55, the encoding method or device or the decoding method or device,

The predetermined value one is 00,

The predetermined value two is 01,

The predetermined value three is 10,

The predetermined value four is 11.

Implementation or Variation 59 (in addition to component conversion, also color space conversion)

In any implementation or variant 1 to 8, 10 to 18, 21 to 27, 29 to 32, 35 to 39, 41, 43 to 55 of the encoding method or device or decoding method or device, the The multi-component is 3 components, and before performing the component conversion forward operation on the 2 components of the 3 components or after performing the component conversion inverse operation, the color space forward transformation is performed on the 3 components or the color space is performed on the 3 components. Spatial inverse transform.

Implementation or variant 60 (in addition to component conversion, also color space conversion)

、

、

变换为R、S、T的重构值

、

、

的逆变换。In the encoding method or device or the decoding method or device described in the implementation or variant 59, the 3 components are R, S, T, and the color space forward transformation is to transform R, S, T into Y, Cg, Co Forward transformation; the color space inverse transformation is to transform, quantize, inversely quantize, and inversely transform the reconstructed values of Y, Cg, and Co, or quantize and inversely quantize the reconstructed values, or transform and inversely transform them. The reconstructed value of

,

,

Transform into reconstructed values of R, S, T

,

,

inverse transformation of .

Implementation or Variation 61 (In addition to component conversion, color space conversion is also performed)

In the encoding method or device or the decoding method or device described in the implementation or variant 60,

The forward transformation that transforms R, S, T into Y, Cg, and Co is:

Y = (2R + S + T)/4

Cg = (2R - S - T)/4

Co = (-S + T)/2;

、

、

变换为

、

、

的逆变换是：said will

,

,

transform to

,

,

The inverse transform of is:

=

+

=

+

=

-

-

=

-

-

=

-

+

；

=

-

+

;

or,

Another simpler form, the forward transformation of R, S, T to Y, Cg, Co is:

Co = T - S

tmp = S + (Co >> 1)

Cg = R - tmp

Y = tmp + (Cg >> 1);

、

、

变换为

、

、

的逆变换是：said will

,

,

transform to

,

,

The inverse transform of is:

– (

>> 1)tmp =

– (

>> 1)

=

+ tmp

=

+ tmp

= tmp –(

>> 1)

= tmp –(

>> 1)

=

+

；

=

+

;

、

、

分别重新命名为新的

、

、

，逆变换就变为：In the above inverse transformation, the

,

,

Renamed separately as new

,

,

, the inverse transformation becomes:

– (

>> 1)tmp =

– (

>> 1)

=

+ tmp

=

+ tmp

= tmp – (

>> 1)

= tmp – (

>> 1)

=

+

。

=

+

.

Claims (10)

1. a method or apparatus for multi-component data encoding, comprising at least the steps or modules of:

1) analyzing the characteristics of the current multi-component residual data, and selecting one of a plurality of preset sets of coefficients for component conversion as a selected coefficient of the current code according to a preset rule;

performing a component transform positive operation on a current residual using at least the selected coefficients;

3) and writing the current coding result into a compressed data code stream, wherein the compressed data code stream at least comprises part or all information called selected information and used for representing which set of coefficients the selected coefficients are.

2. A method or apparatus for multi-component data decoding, comprising at least the steps or modules for performing the following functions and operations:

1) analyzing the compressed data code stream, and at least acquiring part or all information which is called selected information and is used for expressing and selecting which coefficient of a plurality of preset coefficients is selected as a selected coefficient to perform component conversion on the current residual error;

2) selecting one of a plurality of predetermined sets of coefficients as a selected coefficient for current decoding at least according to the information and/or a predetermined parameter and/or a predetermined variable related to decoding;

3) and performing component transformation inverse operation on the residual error at least by using the selected coefficient to obtain restored or reconstructed multi-component residual error data.

3. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, characterized in that the data comprises one or a combination of the following types of data:

one-dimensional data;

two-dimensional data;

multidimensional data;

a graph;

dimension division graphics;

an image;

a sequence of images;

video;

audio frequency;

a file;

a byte;

a bit;

a pixel;

a three-dimensional scene;

a sequence of continuously changing three-dimensional scenes;

a virtual reality scene;

sequence of scenes of continuously changing virtual reality

An image in the form of pixels;

transform domain data of the image;

a set of bytes in two or more dimensions;

a set of bits in two or more dimensions;

a set of pixels;

a set of single component pixels;

a set of three-component pixels (R, G, B, A);

a set of three-component pixels (Y, U, V);

a set of three-component pixels (Y, Cb, Cr);

a set of three-component pixels (Y, Cg, Co);

a set of four component pixels (C, M, Y, K);

a set of four component pixels (R, G, B, A);

a set of four component pixels (Y, U, V, A);

a set of four component pixels (Y, Cb, Cr, A);

a set of four component pixels (Y, Cg, Co, a).

4. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, wherein 2 components out of N (N ≧ 3) components are component-converted, and the remaining N-2 components are not component-converted; the component transformation positive operation in the encoding method or apparatus is a linear transformation F that transforms components w and x into components y and z, with a set of coefficients of I (I ≧ 3); the component conversion in the decoding method or apparatus is to invert a component

And

conversion into components

And

has a coefficient of the set I.

5. The encoding method or apparatus or the decoding method or apparatus according to claim 4,

the linear transformation F has set I coefficients o [ I ], p [ I ], q [ I ], r [ I ], s [ I ], t [ I ], u [ I ], v [ I ], 0 ≦ I < I, the linear transformation F is calculated in the following way:

y = (o[i]*w + p[i]*x + q[i])/r[i]，

z = (s[i]*w + t[i]*x + u[i])/v[i]；

the linear transformation G has set I coefficients a [ I ], b [ I ], c [ I ], d [ I ], e [ I ], f [ I ], G [ I ], h [ I ], 0 ≦ I < I, the linear transformation G being calculated in the following way:

= (a[i]*

+ b[i]*

+ c[i])/d[i]，

= (e[i]*

+ f[i]*

+ g[i])/h[i]。

6. the encoding method or apparatus or the decoding method or apparatus according to claim 5, wherein the I (0 ≦ I < I) th set of coefficients o [ I ], p [ I ], q [ I ], r [ I ], s [ I ], t [ I ], u [ I ], v [ I ], is abbreviated as [ o, p, q, r, s, t, u, v ], the I (0 ≦ I) set of coefficients a [ I ], b [ I ], c [ I ], d [ I ], e [ I ], f [ I ], g [ I ], h [ I ], is abbreviated as [ a, b, c, d, e, f, g, h ], the plurality of sets of coefficients comprising at least some or all of the following seven sets of coefficients:

1) coefficient # 1: [ o, p, q, r, s, t, u, v ] = [1, 0, 0, 0, 0, 1, 0, 0], [ a, b, c, d, e, f, g, h ] = [1, 0, 0, 0, 0, 1, 0, 0] i.e., the linear transformation is an identity transformation, and the formula for the calculation that the component transformation using the present set of coefficients is operating is:

y = w，

z = x；

and the calculation formula of the component conversion inverse operation is

=

，

=

；

2) Coefficient #2 a: [ o, p, q, r, s, t, u, v ] = [1, 1, 2, 1, -1, λ, 2], [ a, b, c, d, e, f, g, h ] = [1, 1, 0, 1, 1, -1, 0, 1], where =0 or 1 or-1 and λ =0 or 1 or-1 are both round-robin control parameters, the calculation formula for the positive operation of component transformation using this set of coefficients is:

y = (w + x +)/2, where =0 or 1 or-1,

z = (w-x + λ)/2, where λ =0 or 1 or-1;

and the calculation formula of the component conversion inverse operation is

=

+

，

=

-

；

3) Coefficient #2 b: [ o, p, q, r, s, t, u, v ] = [1, -1, 2, 1, 1, λ, 2], [ a, b, c, d, e, f, g, h ] = [1, 1, 0, 1, -1, 1, 0, 1], where =0 or 1 and λ =0 or 1 are control parameters in a round-robin fashion, the calculation formula that is operating using component conversion of this set of coefficients is:

y = (w-x +)/2, where =0 or 1 or-1,

z = (w + x + λ)/2, where λ =0 or 1 or-1;

and the calculation formula of the component conversion inverse operation is

=

+

，

= -

+

；

It can be seen that the difference between the coefficient #2b and the coefficient #2a is that p, t, e and f of the two sets of coefficients are opposite numbers to each other;

4) coefficient #3 a: [ o, p, q, r, s, t, u, v ] = [ A, 2, A +1, 1, -2, λ, A +1], [ a, b, c, d, e, f, g, h ] = [1, 1, 0, 1, 1, -A, 2], where A =1 or 4 or other predetermined integer constant satisfying 1 ≦ A ≦ 10, and λ, are round-robin control parameters: a predetermined integer constant of ≦ A/2 or other satisfying-A/2 ≦ A/2, λ =0 or 1 or A/2 or-A/2 or other predetermined integer constant of-A/2 ≦ A/2, =0 or 1 or-1, the calculation formula for the component transform using the present set of coefficients being operated on is:

y = (Aw + 2x +)/(A+1)，

z = (w - 2x +λ)/(A+1)；

and the calculation formula of the component conversion inverse operation is

=

+

，

= (

- A

(+) 2, wherein =0 or 1 or-1;

5) coefficient #3 b: [ o, p, q, r, s, t, u, v ] = [ A, -2, A +1, 1, 2, λ, A +1], [ a, b, c, d, e, f, g, h ] = [1, 1, 0, 1, -1, A, 2], where A =1 or 4 or other predetermined integer constant satisfying 1 ≦ A ≦ 10, and λ, are round-robin control parameters: a predetermined integer constant of ≦ A/2 or other satisfying-A/2 ≦ A/2, λ =0 or 1 or A/2 or-A/2 or other predetermined integer constant of-A/2 ≦ A/2, =0 or 1 or-1, the calculation formula for the component transform using the present set of coefficients being operated on is:

y = (Aw - 2x +)/(A+1)，

z = (w + 2x +λ)/(A+1)；

and the calculation formula of the component conversion inverse operation is

=

+

，

= (-

+ A

(+) 2, wherein =0 or 1 or-1;

it can be seen that the difference between the coefficient #3b and the coefficient #3a is that p, t, e and f of the two sets of coefficients are opposite numbers to each other;

6) coefficient #4 a: [ o, p, q, r, s, t, u, v ] = [2, A +1, -2, 1, λ, A +1], [ a, b, c, d, e, f, g, h ] = [1, -A, 2, 1, 1, 0, 1], where A =1 or 4 or other predetermined integer constant satisfying 1 ≦ A ≦ 10, and λ, are round-robin control parameters: a predetermined integer constant of ≦ A/2 or other satisfying-A/2 ≦ A/2, λ =0 or 1 or A/2 or-A/2 or other predetermined integer constant of-A/2 ≦ A/2, =0 or 1 or-1, the calculation formula for the component transform using the present set of coefficients being operated on is:

y = (2w + Ax +)/(A+1)，

z = (-2w + x +λ)/(A+1)；

and the calculation formula of the component conversion inverse operation is

= (

- A

(+) or 2, where =0 or 1 or-1,

=

+

；

it can be seen that coefficient #4a is the result of the pairs o and p, pairs s and t, pairs a and e, pairs b and f, pairs c and g, and pairs d and h in coefficient #3 a;

7) coefficient #4 b: [ o, p, q, r, s, t, u, v ] = [ -2, A +1, 2, 1, λ, A +1], [ a, b, c, d, e, f, g, h ] = [ -1, A, 2, 1, 1, 0, 1], where A =1 or 4 or other predetermined integer constant satisfying 1 ≦ A ≦ 10, and λ, are round-robin control parameters: a predetermined integer constant of ≦ A/2 or other satisfying-A/2 ≦ A/2, λ =0 or 1 or A/2 or-A/2 or other predetermined integer constant of-A/2 ≦ A/2, =0 or 1 or-1, the calculation formula for the component transform using the present set of coefficients being operated on is:

y = (-2w + Ax +)/(A+1)，

z = (2w + x +λ)/(A+1)；

and the calculation formula of the component conversion inverse operation is

= (-

+ A

(+) or 2, where =0 or 1 or-1,

=

+

；

it can be seen that the difference between the coefficient #4b and the coefficient #4a is that o, s, a and b of the two sets of coefficients are opposite numbers to each other;

it can also be seen that coefficient #4b is the result of the pairs o and p, pairs s and t, pairs a and e, pairs b and f, pairs c and g, and pairs d and h in coefficient #3 b.

7. The encoding method or apparatus or the decoding method or apparatus according to claim 6,

using QP value used for quantizing or de-quantizing the component without component conversion or the component with identity conversion as basic QP_base，

The component y obtained by the non-identity transformation with the ith set of coefficients is quantized or dequantized with a QP value of

QPy[i] = QP_base + DQPy[i]，

The component z obtained by the non-identity transformation with the ith set of coefficients is quantized or dequantized with a QP value of

QPz[i] = QP_base + DQPz[i]，

Wherein each ith set of coefficients has a respective independent QP offset value DQPy [ i ] and DQPz [ i ].

8. The encoding method or apparatus or the decoding method or apparatus according to claim 7,

the value of DQPy [ i ] is-4 or-3 or-2 or-1 or 0 or 1 or 2 or 3;

DQPz [ i ] has a value of DQPy [ i ] or DQPy [ i ] + -1 or DQPy [ i ] + -2 or DQPy [ i ] + -3 or Q, wherein Q ≧ (DQPy [ i ] + 6);

the i-th set of coefficients having a value of DQPz [ i ] of DQPy [ i ] or DQPy [ i ] + -1 or DQPy [ i ] + -2 or DQPy [ i ] + -3 is called the equal component coefficients having equal QP offset values,

and the value with DQPz [ i ] is Q, where the i-th set of coefficients for Q ≧ DQPy [ i ] + 6 is referred to as the primary and secondary component coefficients with the primary and secondary QP offset values, at which time component y and component z are referred to as the primary and secondary components, respectively.

9. The encoding method or apparatus or the decoding method or apparatus according to claim 8, wherein some of the seven sets of coefficients are extended in combination with a QP offset value into the plurality of sets of coefficients used by the following encoding method or apparatus or decoding method or apparatus:

1) i of the seven sets of coefficients₁The set coefficient is combined with the equal QP offset value as I₁The coefficients of the equal components are then smoothed,

2) i of the seven sets of coefficients₂The sleeve coefficient is combined with the primary and secondary QP offset values to I₂The coefficient of the primary and secondary components is sleeved,

whereby the encoding method or apparatus or the decoding method or apparatus uses a total of I₁ + I₂The set of coefficients are subjected to component conversion forward or reverse operations.

10. The encoding method or apparatus or the decoding method or apparatus according to claim 8, wherein the I is₁ + I₂The set of coefficients are divided into predetermined K (2. ltoreq. K.ltoreq.6) sets of coefficients, namely K coefficient sets, each coefficient set having 4 sets of coefficients, the 4 sets of coefficients including a coefficient # 1; a compressed block using component conversion can only use coefficients within one coefficient group: presence of selected information in compressed data code stream called compressed block component conversionRepresents some or all of the information needed for a compressed block to use which of the K coefficient sets.

Images