CN111385555A

CN111385555A - Data compression method and device for inter-component prediction of original and/or residual data

Info

Publication number: CN111385555A
Application number: CN201911405855.1A
Authority: CN
Inventors: 林涛
Original assignee: Tongji University
Current assignee: Tongji University
Priority date: 2018-12-28
Filing date: 2019-12-30
Publication date: 2020-07-07

Abstract

The invention provides a data compression method and a device which adaptively adopt original data component prediction and/or residual data component prediction at the coding and decoding block level. Firstly, a prediction mode including an original data component inter-prediction mode is used for carrying out prediction operation on a coding and decoding block, and the correlation among original values of all components is eliminated to a certain extent; then, for the residual generated after prediction of the coding and decoding block, residual data inter-component prediction is further used to eliminate the correlation between the residual values of the components.

Description

Data compression method and device for inter-component prediction of original and/or residual data

Technical Field

The present invention relates to an encoding and decoding system for lossy or lossless compression of data, and more particularly to a method and apparatus for encoding and decoding multi-component data such as color images and video data.

Background

With the human society entering the era of big data, cloud computing, mobile computing, cloud-mobile computing, ultra high definition (4K) and ultra high definition (8K) video image resolution, 4G/5G communication, and virtual reality, it becomes an indispensable technology to compress various data, including big data, image data, and video data, with ultra high compression ratio and extremely high quality.

A data set is a collection of samples of data, also called data elements (e.g., bytes, bits, pixels, spatial sampling points, transform domain coefficients). When encoding (and corresponding decoding) data compression of a data set (e.g., a file, a one-dimensional data queue, a two-dimensional data file, a frame of image, a video sequence, a transform domain, a transform block, a plurality of transform blocks, a three-dimensional scene, a sequence of three-dimensional scenes that continuously change) arranged in a certain spatial (one-dimensional, two-dimensional, or multi-dimensional) shape and having a certain number of elements and a certain sampling format, especially a two-dimensional or more data sets, the data set is usually divided into a subset of blocks having a predetermined shape and size (i.e., number of elements), called encoding blocks (i.e., decoding blocks from the decoding perspective, collectively referred to as encoding blocks), and the encoding or decoding is performed on a block-by block basis in units of encoding blocks. At any one time, the coding block being coded is referred to as the current coding block. At any one time, the decoding block being decoded is referred to as the current decoding block. The current encoding block or current decoding block is collectively referred to as the current encoding and decoding block or simply the current block.

A data element being encoded or decoded (referred to simply as an element) is referred to as a currently encoded data element or a currently decoded data element, collectively referred to as a current data element, referred to simply as a current element. An element consists of K components (usually 1. ltoreq. K.ltoreq.5), so both the data set and the codec block also consist of K components. For example, elements of one frame image, i.e., pixels, are arranged in a rectangular shape, have a size (resolution) of 1920 (width) x 1080 (height), and are composed of 3 components: g (green), B (blue), R (red) or Y (luminance), U (Cb), V (Cr).

The elements or components thereof are collectively referred to as samples. It is generally determined from the context whether a sample refers to an element or a component of an element. A sample refers to an element and/or a component of an element if it cannot be determined from the context whether the sample refers to the element or a component of the element.

For example, for computer-generated images containing graphics and text, a sampling format called 4:4:4 (444 for short) is generally used, that is, three components of a data set all have the same sampling rate and size (i.e., the number of component samples) for natural images and videos captured by a camera, a sampling format called 4:2:0 (420) is generally used, that is, the sampling rate and size of two components called subcomponents (D components and E components) of a data set having a rectangular shape and three components (e.g., an image or a video) are one quarter of those called another component (F component) in the case of a primary image or video format [ E ] [ 1 j ] [ 1] and E component [ E ] are generally used, that is, if the sampling rate and size of two components called sub-components (D components and E components) of a data set having a rectangular shape and three components [ E ] are generally used, that is, for example, the image or video { E-component { E-sub-1 = [ E } 0-E } 0 = ] [ 1 j ] [ 1, N = ] [ 1 j ] [ 1] (E-sub-0, E component) is generally used, that is, E-component (E-sub-component) and E-sub-component (E-sub-E-component) are generally used, E-sub-E-sub-1, E-sub-component (E-sub-component) in the case of a sampling format (E-sub-1 [ E-sub-component (E-sub-component) of a sub-component (E-sub-component) is generally used, E-sub-component (E-sub-1, E-sub-component) of a sub-component (E-sub-component) of a sub-component (E-sub-component) of a sub-component (E-sub-component) of a sub-component (E-sub-component) of a sub-component (E-sub-component) of a sub-component) of a sub-component (E-sub-component) of a sub-component (E-sub-component) of a sub-component (E-component) of a sub-component) of a sub-.

An image or video in the 444 sample format undergoes 2:1 downsampling in one direction (e.g., horizontal or vertical) of the D and E components to produce a 422 sample format, and undergoes horizontal 2:1 downsampling and vertical 2:1 downsampling (overall 4:1 downsampling) of the D and E components to produce a 420 sample format.

There is generally a degree of correlation between the various components of the multi-component data. Therefore, using one part of the components to predict another part of the components can effectively perform the role of data compression.

Inter-component prediction can be divided into two categories. One is inter-component prediction of the original data, i.e. using reconstructed values of the original data of one or several components to construct predicted values of the original values of other components (i.e. the values of the components themselves), and inter-predicting the other components. The other method is residual data inter-component prediction, namely, each component firstly predicts each component of original data to obtain the residual data of each component, namely the prediction error of the original data, then uses the residual value (namely, the residual data reconstruction value) of one or more components for calculating the original data reconstruction value to construct the prediction value of the residual value (namely, the residual data reconstruction value) of other components for calculating the original data reconstruction value, and carries out inter-component prediction on the residual of other components.

It should be noted that, since the prediction operation is an operation that can be performed multiple times, and a new residual can be generated by performing a prediction on the predicted residual again, the original data and the residual data are relative concepts. For one-time prediction operation, data before the prediction operation is referred to as original data, and prediction difference data obtained after the prediction operation is referred to as residual data. The original data of one prediction operation is allowed to be the residual data of another prediction operation. The residual data of one prediction operation is also allowed to be the original data of another prediction operation.

A reconstructed value of data is the reconstructed data obtained after subjecting said data to encoding operations, typically including prediction and/or transformation and/or quantization, and their inverse. In case the encoding operation and its inverse are lossless, the reconstructed value of the data is equal to the data itself.

In principle, under certain conditions, the original data component inter prediction is equivalent to the residual data component inter prediction. Thus, in the related art, it is considered that the original data component inter-prediction and the residual data component inter-prediction have almost the same effect. Therefore, in the existing image and video codec technology, either original data (i.e., pixel data) component inter prediction is employed instead of residual data component inter prediction for pixel data of an entire image or an entire video sequence, or residual data component inter prediction is employed instead of original data (i.e., pixel data) component inter prediction for pixel data of an entire image or an entire video sequence.

In existing image and video codec techniques, original data component inter-prediction is employed for pixel data of an image or video of 420 sample format, and residual data component inter-prediction is employed for pixel data of an image or video of 444 sample format. In existing image and video coding techniques, neither original nor residual data component inter prediction is employed when using string prediction (also known as string matching or string replication) coding techniques.

The prior art does not consider that in a practical application scenario, since each component may have different prediction modes, different transformation modes, and different quantization factors, the original data component inter-prediction and the residual data component inter-prediction actually have different and complementary effects in some situations, and particularly in the non-444 sampling format, the different and complementary effects are more obvious under certain conditions.

Disclosure of Invention

In order to fully utilize the correlation among the components to improve the coding efficiency, the invention provides a data compression method and a data compression device which adaptively adopts the prediction among original data components and/or adopts the prediction among residual data components at the level of coding and decoding blocks. Firstly, a prediction mode including an original data component inter-prediction mode is used for carrying out prediction operation on a coding and decoding block, and the correlation among original values of all components is eliminated to a certain extent; then, for the residual generated after prediction of the coding and decoding block, residual data inter-component prediction is further used to eliminate the correlation between the residual values of the components.

The primary technical feature of the present invention is that when encoding and decoding data having a plurality of (two or more) components, a part of the encoding and decoding blocks uses at least the original data component inter prediction, another part of the encoding and decoding blocks uses at least the residual data component inter prediction, and another part of the encoding and decoding blocks uses at least both the original data component inter prediction and the residual data component inter prediction.

In the present invention, preferably, the compressed data code stream output by the encoder (i.e. the compressed data code stream input by the decoder) at least contains information indicating whether the coding/decoding block uses the original data inter-component prediction and/or information indicating whether the coding/decoding block uses the residual data inter-component prediction.

In the present invention, it is preferable that the data have three components consisting of one principal component and two secondary components; the data of the three components has a 444 sample format or a 422 sample format or a 420 sample format.

The most basic special technical characteristics of the coding method or device of the invention are that the coding method or device at least comprises a step or module for carrying out interframe predictive coding and/or series predictive coding and/or intraframe predictive coding including the inter-prediction of original data components on original data, and simultaneously comprises a step or module for carrying out residual data component inter-prediction on residual data generated after predictive coding, and the output compressed data code stream at least comprises information which indicates whether the coding block uses the inter-prediction of the original data components and/or uses the inter-prediction of the residual data components. Fig. 1 is a schematic diagram of an encoding method or apparatus of the present invention.

The most basic special technical characteristic of the decoding method or the device is that a compressed data code stream is analyzed, and at least information which indicates whether a decoding block uses original data component inter-prediction and/or residual data component inter-prediction is obtained; if the information indicates that the decoded block uses inter-component prediction of the original data, constructing predicted values of the original data of other components from at least reconstructed values of the original data of one or several components; if the information indicates that the decoded block uses inter-component prediction of residual data, then a prediction value for the residual value used to calculate its original data reconstruction value (i.e., residual data reconstruction value) for the other component is constructed from at least the residual value of one or several components used to calculate its original data reconstruction value (i.e., residual data reconstruction value). Fig. 2 is a schematic diagram of a decoding method or apparatus of the present invention.

According to an aspect of the present invention, there is provided an encoding method or apparatus for compressing multi-component data, comprising at least one step or module for performing one or a combination of the following functions and operations:

inter-prediction encoding and/or series prediction encoding and/or intra-prediction encoding of original data including inter-prediction of original data components, the inter-prediction of original data components including at least prediction values and prediction difference values of original data or reconstruction values thereof constructing Q (Q ≧ 1) components from original data or reconstruction values thereof of at least H (H ≧ 1) components, wherein H is allowed to be equal to Q, some or all of the H components are allowed to overlap with some or all of the Q components, and the prediction difference values of one or more of the Q components are allowed to be zero;

performing residual coding including residual data component prediction on residual data generated after predictive coding, wherein the residual data component prediction at least comprises a prediction value and a prediction difference value, wherein the prediction value and the prediction difference value of the residual data of Z (Z is more than or equal to 1) components are constructed from the residual data of at least W (W is more than or equal to 1) components or the reconstruction values of the residual data, the W is allowed to be equal to Z, part or all of the W components are allowed to be overlapped with part or all of the Z components, and the prediction difference value of one or more of the Z components is allowed to be zero;

at least information indicating whether the coding block uses the original data component inter prediction and/or uses the residual data component inter prediction is written into the compressed data code stream.

From a first aspect, the present invention provides an encoding method for compressing multi-component data, characterized by at least one or a combination of the following steps:

step 1) inter-frame prediction coding and/or series prediction coding and/or intra-frame prediction coding including original data component prediction are carried out on original data, wherein the operation of the original data component prediction at least comprises the steps of constructing predicted values of the original data of Q (Q is larger than or equal to 1) components or reconstructed values thereof and prediction difference values from the original data of at least H (H is larger than or equal to 1) components or reconstructed values thereof, wherein H is allowed to be equal to Q, part or all of the H components are allowed to be overlapped with part or all of the Q components, and the prediction difference values of one or more of the Q components are allowed to be zero;

step 2) residual coding including residual data component inter-prediction is carried out on residual data generated after predictive coding, wherein the residual data component inter-prediction operation at least comprises the steps of constructing a predicted value and a predicted difference value of the residual data of Z (Z ≧ 1) components or a reconstructed value thereof from the residual data of at least W (W ≧ 1) components or reconstructed values thereof, wherein W is allowed to be equal to Z, part or all of the W components are allowed to be overlapped with part or all of the Z components, and the predicted difference value of one or more of the Z components is allowed to be zero;

and 3) at least writing information which indicates whether the coding block uses the original data component inter-prediction and/or uses the residual data component inter-prediction into the compressed data code stream.

From a second aspect, the present invention provides an encoding apparatus for compressing multi-component data, characterized by comprising at least one or a combination of the following modules:

1) a predictive coding module: inter-prediction encoding and/or series prediction encoding and/or intra-prediction encoding of raw data including inter-raw data component prediction, the operation of the inter-raw data component prediction at least including constructing a prediction value of raw data of Q (Q ≧ 1) components or a reconstruction value thereof and a prediction difference value from raw data of at least H (H ≧ 1) components or a reconstruction value thereof, wherein H is allowed to be equal to Q, some or all of the H components are allowed to overlap with some or all of the Q components, and the prediction difference value of one or more of the Q components is allowed to be zero;

2) a residual coding module: residual coding including residual data component inter-prediction is carried out on residual data generated after the residual data are subjected to the prediction coding, wherein the residual data component inter-prediction at least comprises the steps of constructing a predicted value and a prediction difference value of the residual data of Z (Z ≧ 1) components or the reconstruction value of the residual data of the Z (Z ≧ 1) components from the residual data of at least W (W ≧ 1) components or the reconstruction value of the residual data of the Z components, wherein W is allowed to be equal to Z, part or all of the W components are allowed to be overlapped with part or all of the Z components, and the prediction difference value of one or more of the Z components is allowed to be zero;

3) a compressed data code stream generation module: at least information indicating whether the coding block uses the original data component inter prediction and/or uses the residual data component inter prediction is written into the compressed data code stream.

According to another aspect of the present invention, there is provided a decoding method or apparatus for compressing multi-component data, comprising at least one step or module for performing one or a combination of the following functions and operations:

analyzing the compressed data code stream, and at least acquiring information which indicates whether the decoding block uses original data component inter-prediction and/or residual data component inter-prediction;

performing raw data inter-component prediction, wherein the raw data inter-component prediction at least comprises constructing predicted values of raw data of Q (Q ≧ 1) components or reconstructed values thereof from raw data of at least H (H ≧ 1) components or reconstructed values thereof and/or constructing raw data reconstructed values of the Q components from the predicted values of the Q components and/or predicted difference values of the Q components or reconstructed values thereof, wherein the H is allowed to be equal to the Q, some or all of the H components are allowed to overlap with some or all of the Q components, and the predicted difference values or reconstructed values thereof of one or more of the Q components are allowed to be zero;

performing residual data inter-component prediction, the residual data inter-component prediction at least comprising constructing predicted values of residual data of Z (Z ≧ 1) components or reconstructed values thereof from residual data of at least W (W ≧ 1) components or reconstructed values thereof and/or constructing residual data reconstructed values of the Z components from the predicted values of the Z components and/or predicted difference values of the Z components or reconstructed values thereof, wherein allowing the W to be equal to the Z, allowing some or all of the W components to overlap with some or all of the Z components, and allowing the predicted difference values of one or more of the Z components or reconstructed values thereof to be zero.

From a third perspective, the present invention provides a decoding method for compressing multi-component data, characterized by at least one or a combination of the following steps:

step 1) analyzing a compressed data code stream, and at least acquiring information which indicates whether a decoding block uses original data component inter-prediction and/or residual data component inter-prediction;

step 2) performing original data inter-component prediction, wherein the operation of the original data inter-component prediction at least comprises constructing predicted values of original data of Q (Q ≧ 1) components or reconstructed values thereof from original data of at least H (H ≧ 1) components or reconstructed values thereof and/or constructing original data reconstructed values of the Q components from the predicted values of the Q components and/or predicted difference values of the Q components or reconstructed values thereof, wherein H is allowed to be equal to Q, some or all of the H components are allowed to overlap with some or all of the Q components, and the predicted difference values or reconstructed values thereof of one or more of the Q components are allowed to be zero;

and 3) performing residual data inter-component prediction, wherein the residual data inter-component prediction at least comprises the steps of constructing a predicted value of the residual data of Z (Z ≧ 1) components or a reconstructed value thereof from the residual data of at least W (W ≧ 1) components or the reconstructed value thereof and/or constructing a residual data reconstructed value of the Z components from the predicted value of at least the Z components and/or the predicted difference of the Z components or the reconstructed value thereof, wherein W is allowed to be equal to Z, part or all of the W components are allowed to be overlapped with part or all of the Z components, and the predicted difference of one or more of the Z components or the reconstructed value thereof is allowed to be zero.

From a fourth perspective, the present invention provides a decoding apparatus for compressing multi-component data, characterized by at least one or a combination of the following modules:

1) compressed data code stream analysis module: analyzing the compressed data code stream, and at least acquiring information which indicates whether the decoding block uses original data component inter-prediction and/or residual data component inter-prediction;

2) a predictive decoding module: performing raw data inter-component prediction, wherein the operation of the raw data inter-component prediction at least comprises constructing predicted values of the raw data of Q (Q ≧ 1) components or reconstructed values thereof from the raw data of H (H ≧ 1) components or reconstructed values thereof and/or constructing raw data reconstructed values of the Q components from the predicted values of the Q components and/or predicted difference values of the Q components or reconstructed values thereof, wherein the H is allowed to be equal to the Q, some or all of the H components are allowed to overlap with some or all of the Q components, and the predicted difference values or reconstructed values thereof of one or more of the Q components are allowed to be zero;

3) a residual decoding module: and if the information indicates that the decoded block uses residual data inter-component prediction, performing residual data inter-component prediction on the decoded block, wherein the residual data inter-component prediction at least comprises constructing a predicted value of the residual data of Z (Z ≧ 1) components or a reconstructed value thereof from the residual data of W (W ≧ 1) components or the reconstructed value thereof and/or constructing a residual data reconstructed value of the Z components from the predicted value of the Z components and/or the predicted difference values of the Z components or the reconstructed values thereof, wherein allowing the W to be equal to the Z, allowing part or all of the W components to overlap with part or all of the Z components, and allowing the predicted difference values of one or more of the Z components or the reconstructed values thereof to be zero.

The present invention is applicable to encoding and decoding for lossy compression of data, and is also applicable to encoding and decoding for lossless compression of data. The invention is applicable to encoding and decoding of image data, and the invention is equally applicable to encoding and decoding of any other one-dimensional, two-dimensional or multi-dimensional data.

In the present invention, the data includes one or a combination of the following types of data

1) One-dimensional data;

2) two-dimensional data;

3) multidimensional data;

4) a graph;

5) dimension division graphics;

6) an image;

7) a sequence of images;

8) video;

9) a three-dimensional scene;

10) a sequence of continuously changing three-dimensional scenes;

11) a virtual reality scene;

12) sequence of scenes of continuously changing virtual reality

13) An image in the form of pixels;

14) transform domain data of the image;

15) a set of bytes in two or more dimensions;

16) a set of bits in two or more dimensions;

17) a set of pixels;

18) a set of pixel components.

In the present invention, in the case where data is generated from an image, a sequence of images, a video, or the like, an encoded block or a decoded block is an encoded region or a decoded region of an image, including at least one of: a full picture, a sub-picture of a picture, a slice, a slice tile, a macroblock, a largest coding unit LCU, a coding tree unit CTU, a coding unit CU, a sub-region of a CU, a sub-coding unit SubCU, a sub-region of a prediction unit PU, a sub-prediction unit SubPU, a transform unit TU, a sub-region of a TU, a sub-transform unit SubTU.

In the encoding method or apparatus or the decoding method or apparatus of the present invention, the information indicating whether the encoded and decoded block uses the original data component inter prediction and/or the residual data component inter prediction exists at one or several of the following places of the compressed data code stream:

1) a sequence parameter set; usually one or several directly existing or implicitly derived or mixed syntax elements of the sequence parameter set;

2) a set of image parameters; usually one or several directly existing or implicitly derived or mixed syntax elements of the picture parameter set;

3) a sequence header; one or several syntax elements, either directly or implicitly derived or mixed, usually of the sequence header;

4) an image head; one or several syntax elements, usually either directly or implicitly derived or mixed, of the picture header;

5) a tape head; one or several directly existing or implicitly derived or mixed syntax elements, usually slice headers;

6) a CTU head; one or several syntax elements, either directly or implicitly derived or mixed, typically of the CTU header;

7) a CU header; one or several directly existing or implicitly derived or mixed syntax elements, usually CU headers;

8) encoding and decoding a block header; one or several directly existing or implicitly derived or mixed syntax elements, typically a codec block header;

the directly-existing syntax element consists of one or more bit strings (bit strings) in the compressed data code stream. The implicitly derived syntax element is a syntax element derived from other coding parameters and/or codec variables and/or other syntax elements of the compressed data stream or a predetermined syntax element default value. The mixed syntax element is a syntax element that is partly directly present (i.e. composed of one or more bit strings in the compressed data stream) and partly implicitly derived (i.e. derived from other coding parameters and/or codec variables and/or other syntax elements of the compressed data stream and/or predetermined syntax element default values).

The technical features of the present invention are explained above by specific embodiments. Other advantages and effects of the present invention will be readily apparent to those skilled in the art from the disclosure herein. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention.

Drawings

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

Fig. 2 is a schematic diagram of a decoding method or apparatus of the present invention.

Fig. 3 is a positional relationship of a sub-component sampling point with respect to a main-component sampling point in four downsampling types called type 0, type 1, type 2, and type 3, respectively.

Detailed Description

The following are further implementation details or variations of the present invention.

The encoding method or apparatus or the decoding method or apparatus according to embodiment 1, wherein the data is an image or video having three components, namely, a primary component and two secondary components.

The encoding method or apparatus or the decoding method or apparatus according to embodiment 2, wherein the data is an image or video having three components, namely, a primary component and two secondary components, and the three components are one of the following cases:

y luminance component, U chrominance component, V chrominance component, or Y luminance component, Cb chrominance component, Cr chrominance component, or R red component, G green component, B blue component, or G green component, R red component, B blue component, or G green component, B blue component, R red component.

Example 3

In the encoding method or apparatus or the decoding method or apparatus, the data is an image or video having three components of one primary component and two secondary components, original data and/or residual data are simply referred to as original/residual data, and in the inter-component prediction of the original/residual data,

h-1 and Q-2 and/or W-1 and Z-2, predicting the two secondary components using the primary component; in the encoding method or apparatus, constructing a prediction value and a prediction difference value of the original/residual data of the two secondary components or the reconstruction values thereof from the original/residual data of at least the primary component or the reconstruction values thereof; in a decoding method and apparatus, original/residual data reconstruction values for at least two secondary components are constructed from prediction difference values or reconstruction values thereof for the two secondary components

And/or

H1 and Q1 and/or W1 and Z1, predicting one of the two secondary components using the other of the two secondary components; in the encoding method or apparatus, a prediction value and a prediction difference value of original/residual data of one secondary component of the two secondary components or a reconstruction value thereof are constructed from the original/residual data of the other secondary component of the two secondary components or the reconstruction value thereof; in the decoding method and apparatus, a reconstructed value of original/residual data of one of the two sub-components is constructed from a predicted difference value of the one of the two sub-components or a reconstructed value thereof

And/or

H2 and Q2 and/or W2 and Z2, predicting the other of the two secondary components using either of the two secondary components; in the encoding method or apparatus, constructing a prediction value and a prediction difference value of the original/residual data or the reconstructed values of the two secondary components from the original/residual data or the reconstructed values of the at least two secondary components; in the decoding method and apparatus, original/residual data reconstruction values of at least the two secondary components are constructed from prediction difference values of the two secondary components or reconstruction values thereof.

Example 4

In the encoding method or apparatus or the decoding method or apparatus, the data is an image or video in a 422-sample format having three components, namely, one primary component and two secondary components, the sampling rate and size of the two secondary components being one-half of the sampling rate and size of the primary component, respectively, the two secondary components are predicted using the primary component in the original data component inter-prediction and the residual data component inter-prediction, and both the operation of the original data component inter-prediction and the operation of the residual data component inter-prediction include an operation of horizontal 2:1 down-sampling or vertical 2:1 down-sampling the primary component.

Example 5

In the encoding method or apparatus or the decoding method or apparatus, the data is an image or video of a 420-sample format having three components, one primary component and two secondary components, the sampling rate and size of the two secondary components being one quarter of the sampling rate and size of the primary component, respectively, the two secondary components are predicted using the primary component in the inter-raw data component prediction and the inter-residual data component prediction, and the operation of the inter-raw data component prediction and the operation of the inter-residual data component prediction both include operations of horizontal 2:1 down-sampling and vertical 2:1 down-sampling of the primary component in total 4:1 down-sampling.

Example 6

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 5, the 4:1 downsampling is to calculate a weighted average value of I × J principal components as a value of the downsampled principal components according to one of the following downsampling formulas:

downsampling equation 1 calculation of the value of the downsampled principal component from the values P (2i,2j), P (2i +1,2j), P (2i,2j +1), P (2i +1,2j +1) of 2 × 2 principal components

P₄₂₀(i, j) = ((P(2i,2j)+P(2i+1,2j)+P(2i,2j+1)+P(2i+1,2j+1)+2)>>2, wherein>>Is a 2-bit right shift;

downsampling equation 2 the value of the downsampled principal component is calculated from the values P (2i-1,2j), P (2i,2j), P (2i +1,2j), P (2i-1,2j +1), P (2i,2j +1), P (2i +1,2j +1) of 3 × 2 principal components

P₄₂₀(i, j) = (P(2i-1,2j)+2P(2i,2j)+P(2i+1,2j)+P(2i-1,2j+1)+2P(2i,2j+1)+P(2i+1,2j+1)+4)>>3；

Downsampling equation 3 the value of the downsampled principal component is calculated from the values P (2i +1,2j), P (2i,2j), P (2i +1,2j), P (2i +2,2j), P (2i-1,2j +1), P (2i,2j +1), P (2i +1,2j +1), P (2i +2,2j +1) of 4 × 2 principal components

P₄₂₀(i, j) = (P(2i-1,2j)+3P(2i,2j)+3P(2i+1,2j)+P(2i+2,2j)+P(2i-1,2j+1)+3P(2i,2j+1)+3P(2i+1,2j+1)+ P(2i+2,2j+1)+8)>>4；

Downsampling equation 4 from the values P (2i-1,2j-1), P (2i,2j-1), P (2i +1,2j-1), P (2i-1,2j), P (2 i-2 j), P (2i,2j), P (2i +1,2j), P (2i-1,2j +1), P (2i,2j +1), P (2i +1,2j +1) of 3 × 3 principal components, the downsampled principal component values are calculated

P₄₂₀(i, j) = (P(2i-1,2j-1)+2P(2i,2j-1)+P(2i+1,2j-1)+2P(2i-1,2j)+4P(2i,2j)+2P(2i+1,2j)+P(2i-1,2j+1)+ 2P(2i,2j+1)+P(2i+1,2j+1)+8)>>4；

Downsampling equation 5 from the values P (2i,2j-1), P (2i +1,2j-1), P (2i,2j), P (2i +1,2j), P (2i,2j +1), P (2i +1,2j +1) of 2 × 3 principal components

P₄₂₀(i, j) = ((P(2i,2j-1)+P(2i+1,2j-1)+2P(2i,2j)+2P(2i+1,2j)+P(2i,2j+1)+P(2i+1,2j+1)+4)>>3。

What downsampling formula to use is generally determined by the positional relationship of the secondary component sample points relative to the primary component sample points in a particular downsampling type. Fig. 3 is a positional relationship of a sub-component sampling point with respect to a main-component sampling point in four downsampling types called type 0, type 1, type 2, and type 3, respectively.

Example 7

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 5, information specifying that one of a plurality of predetermined downsampling formulas is selected to perform downsampling on the principal component exists in the compressed data code stream.

Example 8

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 7, the information specifying that the downsampling operation is performed on the principal component by selecting one of several predetermined downsampling formulas exists at one or several of the following places of the compressed data code stream:

Example 9

Embodiment 8 wherein the one or more directly or implicitly derived or mixed syntax elements and their semantics are

Down-sampling type downsamplin _ type

A type of downsampling and a corresponding downsampling formula are specified. If there is no downsamplin _ type in the compressed data stream, the value of downsamplin _ type is implicitly derived as 0.

Example 10

In the encoding method or apparatus or the decoding method or apparatus, the data is an image or a video, and the information indicating whether the encoded and decoded block uses original data component inter prediction and/or uses residual data component inter prediction and the corresponding operations include the following flag bits and/or identification codes and associated syntax elements in italic representation, directly or indirectly mixed with each other, and the corresponding operations:

sub-component prediction mode flag and/or identification code

If the value of the secondary component prediction mode flag bit and/or the identification code indicates that the current coding and decoding block uses inter-primary data component prediction, then

{

Constructing primary data prediction values of secondary components from primary data reconstruction values of primary components of a current codec block

}

Prediction mode flag and/or identification code

If the value of the prediction mode flag bit and/or the identification code indicates that the current coding and decoding block uses a number of predetermined prediction modes

{

Constructing a prediction value of a residual data reconstruction value of a secondary component from a residual data reconstruction value of a primary component of a current coded/decoded block

}

The direct flag and/or identification code is composed of one or more bit strings (bit strings) in the compressed data code stream. The indirect flag and/or identification code is a flag and/or identification code derived from other coding parameters and/or codec variables and/or other syntax elements of the compressed data stream. The flag bits and/or identification codes of the direct and indirect mixing are flag bits and/or identification codes of partial direct (i.e. composed of one or more bit strings in the compressed data code stream) and partial indirect (i.e. derived from other encoding parameters and/or codec variables and/or other syntax elements of the compressed data code stream) mixing.

Example 11

The encoding method or apparatus or the decoding method or apparatus according to embodiment 10, wherein the plurality of predetermined prediction modes at least include 1) inter prediction modes, and 2) intra prediction modes in which the secondary component direct intra prediction mode (i.e., the secondary component prediction mode) directly takes (i.e., is identical to) the primary component.

Example 12

The encoding method or apparatus or the decoding method or apparatus according to embodiment 10, wherein the plurality of predetermined prediction modes at least include 1) an inter-prediction mode, 2) a secondary component direct intra-prediction mode, i.e., a secondary component prediction mode, which directly takes (i.e., is identical to) an intra-prediction mode of the primary component, and 3) a secondary component prediction mode using inter-prediction of the original data component.

Example 13

In the encoding method or apparatus or the decoding method or apparatus, the data is an image or video having three components, one primary component and two secondary components; reconstructing values from residual data of the principal component, either a 2:1 down-sampled version in the case of the 422 sample format or a 4:1 down-sampled version in the case of the 420 sample format, are denoted as recR_LConstructing a predicted value predR of a residual data reconstruction value of two sub-components D and E, denoted as C = (D, E)_C= (predR_D, predR_E) Comprises at least the following operations expressed by the following formulas:

predR_C= a_C*recR_L，

wherein a is_C= (a_D, a_E) Is a two-component multiplication factor, and the two sub-components D and E have respective multiplication factors a_DAnd a_E。

Example 14

Embodiment 13 the method or apparatus for encoding or decoding as described in the method or apparatus for encoding or decoding, wherein the predictor predR of the residual data reconstruction value of the secondary component is used_CThe prediction difference R of the residual error data reconstruction value of the secondary component_C= (∆R_D, ∆R_E) Constructing a residual data reconstruction value recR of the secondary component expressed by the following formula_C= (recR_D, recR_E)：

recR_C＝∆R_C＋predR_C。

Example 15

Embodiment 13 the encoding method or apparatus or the decoding method or apparatus, wherein the multiplication factor a_DAnd a_EThe value range of (a) is one of the following sets:

R = {

: -2ⁿ⁺¹≤ m ≤ 2ⁿ⁺¹}={0, ±

: 1 ≤ m ≤ 2ⁿ⁺¹} = {±

: 0 ≤ m ≤ 2ⁿ ⁺¹or a subset of R, where n is a predetermined positive number and m is-2ⁿ⁺¹And 2ⁿ⁺¹All or a part of the whole integers or integers therebetween

Or

A subset of R, comprising at least:

R_n= {0,

: m=±2^k, m=±(2ⁿ+2^k) K is 0. ltoreq. n or R_nIncluding the subset 0,

: m=±2^k0 ≦ k ≦ n } or a subset thereof

Or

When n =6, the number of the bits is increased,

R₆={0, ±0.015625, ±0.03125, ±0.0625, ±0.125, ±0.25, ±0.5, ±1, ±1.015625, ±1.03125, ±1.0625, ±1.125, ±1.25, ±1.5, ±2}={0, ±

, ±

,±

, ±

, ±

, ±

, ±1, ±1

, ±1

, ±1

, ±1

, ±1

, ±1

, + -2 } or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

, ±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

, ±

, ±1}={0,±

, ±

, ±

, ±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±1}={0,±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

}={0, ±

, ±

, ±

, ±

Or a subset thereof

Or

When n =1, the number of the bits is set to n =1,

R₁= {0,

: m=±2^k, m=±(2+2^k), 0 ≤ k ≤ 1} = {0,

m = ± 1, ± 2, ± 3, ± 4} or a subset thereof

Or

R₁A subset of {0, +/-

, + -1, + -2 } or a subset thereof

Or

R₁A subset of {0, +/-

, + -1} or a subset thereof

Or

R₁Or a subset thereof {0, 1} or {0, -1 }.

Example 16

Embodiment 13 the encoding method or apparatus or the decoding method or apparatus, wherein the multiplication factor a_DAnd a_EThe value range of (A) is R = {0, +/-

: 1 ≤ m ≤ 2ⁿ⁺¹} = {±

: 0 ≤ m ≤ 2ⁿ⁺¹Either R or a subset of R when a_X=±

Wherein X is D or E, a_XIs a_DOr a_EWhen is, predR_XAnd predR_XIs one of the following predetermined conditions:

predR_X= a_X*recR_L= ±((m*recR_L+ 2^n-1)>>n) wherein>>n is an n-bit (bit) operation of a right-shifted binary,

or

predR_X= a_X*recR_L= ±((m*recR_L)>>n)，

Or

predR_X= a_X*recR_L= (((±m)*recR_L)>>n)。

Example 17

Embodiment 13 the encoding method or apparatus or the decoding method or apparatus, wherein the multiplication factor a_DAnd a_EThe value range of (a) is a function of the size and/or width and/or height of the codec block in which the multiplication factor is located.

Example 18

Embodiment 13 in the encoding method or apparatus or the decoding method or apparatus, the multiplication factor a_DAnd a_EThe values of (a) and (b) are synchronized, i.e. have equal values, between the encoder and the decoder by direct means or by implicit derivation or by hybrid means;

the direct way is to represent the multiplicative factor a by a syntax element composed of one or more bit strings (bit strings) directly present in the compressed data code stream_DAnd a_EA value of (d);

the implicit derivation being from at least the multiplicative factor a_DAnd a_EThe values of the multiplication factor a are obtained by calculation and derivation of other coding parameters and/or coding and decoding variables which have no direct relation and/or syntax elements in compressed data code streams or preset default values of the syntax elements_DAnd a_EA value of (d);

the hybrid approach is an approach that mixes the direct approach and the implicitly derived approach.

Example 19

The encoding method or apparatus or the decoding method or apparatus according to embodiment 18,

the syntax elements in the direct mode include

Representing said multiplication factor a_DAnd/or a_ESyntax element of whether or not zero

And/or

Representing said multiplication factor a_DAnd/or a_EPositive and negative syntax elements of (1), namely sign bit syntax elements

And/or

Representing said multiplication factor a_DAnd/or a_ESyntax element of absolute value of.

Example 20

the implicit derivation mode at least uses the following coding and decoding variables to calculate, derive and obtain the multiplication factor a_DAnd/or a_EThe value of (c):

residual data reconstruction value of principal component of current coding and decoding block

And/or

Residual data reconstruction values for principal components of directly neighboring and/or indirectly neighboring pixels of a current codec block

And/or

Residual data reconstruction values for sub-components of directly neighboring and/or indirectly neighboring pixels of the current codec block.

Example 21

the mixing mode has at least the following characteristics:

representing said multiplication factor a_DAnd a_EOne part of information of the value of (A) directly exists in a compressed data code stream, the other part of information is obtained by an implicit derivation mode, and the multiplication factor a is obtained from at least two parts of information_DAnd a_EThe value of (c).

Example 22

the mixing mode has at least the following characteristics:

denotes a_XWherein X is D or E is a_XIs a_DOr a_EInitial value of ia_XIs directly present in the compressed data stream, from said initial value ia_XAnd/or at least with said a_XThe values of the coding parameters and/or the coding and decoding variables and/or the syntax elements in the compressed data code stream or the default values of the predetermined syntax elements are not directly related, and the a is obtained by calculation and derivation_XThe value of (c).

Example 23

the mixing mode has at least the following characteristics:

denotes a_XWherein X is D or E is a_XIs a_DOr a_EInitial value of ia_XIs directly present in the compressed data stream, from said initial value ia_XAnd/or at least the following coding and decoding variables are calculated, derived and obtained_XThe value of (c):

And/or

Example 24

the mixing mode has at least the following characteristics:

denotes a_XWherein X is D or E is a_XIs a_DOr a_EInitial value of ia_XThe information of (a) exists in the compressed data code stream directly_XIs equal to ia_X+ da_XTherein da_XCalculating, deriving, obtaining from at least the following codec variables:

And/or

Example 25

Embodiment 18 as described above, wherein the direct mode is a mode for encoding or decoding_DAnd a_EPartial combination or all combination of (a)_D, a_E) The value of (2) is subjected to joint binarization, inverse binarization, entropy coding and entropy decoding.

Example 26

Embodiment 18 as described above, wherein the direct mode is a mode for encoding or decoding_DThe binarization, inverse binarization, entropy encoding, entropy decoding of the value of (a) is dependent on a_EValue or pair a of_EThe binarization, inverse binarization, entropy encoding, entropy decoding of the value of (a) is dependent on a_DThe value of (c).

Example 27

Embodiment 18 as described above, wherein the direct mode is a mode for encoding or decoding_DIs zero and a_EThe case of zero is represented by a binary codeword or flag bit (usually 0 or 1) with a code length of one bit.

Example 28 (version on "original data" corresponding to example 13 on "residual data")

In the encoding method or apparatus or the decoding method or apparatus, the data is an image or video having three components, one primary component and two secondary components; reconstructing values from raw data of principal components either in the case of the 422 sample format, which is 2:1 down-sampled version, or in the case of the 420 sample format, which is 4:1 down-sampledThe sampled versions, all denoted recP_LConstructing a predicted value predP of the reconstructed value of the original data of the two subcomponents D and E, denoted as C = (D, E)_C= (predP_D, predP_E) Comprises at least the following operations expressed by the following formulas:

predP_C= B_C+ A_C*recP_L，

wherein B is_C= (B_D, B_E) Is a two component offset, A_C= (A_D, A_E) Is a two-component multiplication factor, and the two sub-components D and E have respective offsets B_DAnd B_EAnd/or respective multiplication factors A_DAnd A_E。

Example 29 (version on "original data" corresponding to example 14 on "residual data")

Embodiment 28 the method or apparatus for encoding or decoding as described above, wherein the prediction value predP of the original data reconstruction value of the secondary component is used_CThe prediction difference value Δ P of the original data reconstruction value of the secondary component_C= (∆P_D, ∆P_E) Constructing a recP value of the original data of the secondary component expressed by the following formula_C= (recP_D, recP_E)：

recP_C＝∆P_C＋predP_C。

Example 30

Embodiment 29 encoding method or apparatus or decoding method or apparatus using both inter-component prediction of original data and inter-component prediction of residual data, where the difference between the original data reconstruction value Δ P_CIs the residual data reconstruction value recR_COne way of calculating the raw data reconstruction value is as follows:

recP_C＝∆P_C＋predP_C＝recR_C＋predP_C＝∆R_C＋predR_C＋predP_C，

wherein, Δ R_CAnd predR_CRespectively, prediction difference values of residual data reconstruction values in inter-residual data component predictionAnd a prediction value of the residual data reconstruction value.

Example 31 (version on "original data" corresponding to example 15 on "residual data")

Embodiment 28 the encoding method or apparatus or the decoding method or apparatus, wherein the multiplication factor A_DAnd A_EThe value range of (a) is one of the following sets:

R = {

: -2ⁿ⁺¹≤ m ≤ 2ⁿ⁺¹}={0, ±

: 1 ≤ m ≤ 2ⁿ⁺¹} = {±

Or

A subset of R, comprising at least:

R_n= {0,

: m=±2^k, m=±(2ⁿ+2^k) K is 0. ltoreq. n or R_nIncluding the subset 0,

: m=±2^k0 ≦ k ≦ n } or a subset thereof

Or

When n =6, the number of the bits is increased,

R₆={0, ±0.015625, ±0.03125, ±0.0625, ±0.125, ±0.25, ±0.5,±1, ±1.015625, ±1.03125, ±1.0625, ±1.125, ±1.25, ±1.5, ±2}={0, ±

, ±

,±

, ±

, ±

, ±

, ±1, ±1

, ±1

, ±1

, ±1

, ±1

, ±1

, + -2 } or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

, ±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

, ±

, ±1}={0,±

, ±

, ±

, ±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±1}={0,±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

}={0, ±

, ±

, ±

, ±

Or a subset thereof

Or

When n =1, the number of the bits is set to n =1,

R₁= {0,

: m=±2^k, m=±(2+2^k), 0 ≤ k ≤ 1} = {0,

m = ± 1, ± 2, ± 3, ± 4} or a subset thereof

Or

R₁A subset of {0, +/-

, + -1, + -2 } or a subset thereof

Or

R₁A subset of {0, +/-

, + -1} or a subset thereof

Or

R₁Or a subset thereof {0, 1} or {0, -1 }.

Example 32 (version on "original data" corresponding to example 16 on "residual data")

Embodiment 28 the encoding method or apparatus or the decoding method or apparatus, wherein the multiplication factor A_DAnd A_EThe value range of (A) is R = {0, +/-

: 1 ≤ m ≤ 2ⁿ⁺¹} = {±

: 0 ≤ m ≤ 2ⁿ⁺¹Either R or a subset of R when A is_X=±

Wherein X is D or E, A_XIs A_DOr A_ETime, predP_XAnd predP_XIs one of the following predetermined conditions:

predP_X= B_X+ A_X*recP_L= B_X± ((m*recP_L+ 2^n-1)>>n) wherein>>n is an n-bit (bit) operation of a right-shifted binary,

or

predP_X= B_X+ A_X*recP_L= B_X± ((m*recP_L)>>n)，

Or

predP_X= B_X+ A_X*recP_L= B_X+ (((±m)*recP_L)>>n)。

Example 33 (version on "original data" corresponding to example 17 on "residual data")

Embodiment 28 the encoding method or apparatus or the decoding method or apparatus, wherein the multiplication factor A_DAnd A_EThe value range of (a) is a function of the size and/or width and/or height of the codec block in which the multiplication factor is located.

Example 34

Embodiment 28 the encoding method or apparatus or the decoding method or apparatus, wherein A_D、A_E、B_D、B_EThe values of (a) and (b) are synchronized, i.e. have equal values, between the encoder and the decoder by direct means or by implicit derivation or by hybrid means;

the direct way is to represent the A bit strings by syntax elements consisting of one or more bit strings (bit strings) directly present in the compressed data code stream_D、A_E、B_D、B_EA value of (d);

the implicit derivation is from at least the same as A_D、A_E、B_D、B_EThe values of the A are obtained by calculation and derivation of other coding parameters and/or coding and decoding variables which have no direct relation and/or syntax elements in the compressed data code stream or preset default values of the syntax elements_D、A_E、B_D、B_EA value of (d);

Example 35

The encoding method or apparatus or the decoding method or apparatus according to embodiment 34,

the implicit derivation mode at least uses the following coding and decoding variables to calculate, derive and obtain the offset B_DAnd/or B_EAnd/or the multiplication factor A_DAnd/or A_EThe value of (c):

raw data reconstruction values for principal components of a current codec block

And/or

Raw data reconstruction values of principal components of directly adjacent and/or indirectly adjacent pixels of a current codec block

And/or

The original data reconstruction values of the sub-components of directly adjacent and/or indirectly adjacent pixels of the current codec block.

Example 36

the mixing mode has at least the following characteristics:

represents said A_DAnd/or A_EAnd/or B_DAnd/or B_EOne part of information of the value (A) directly exists in a compressed data code stream, the other part of information is obtained in an implicit derivation mode, and the A is obtained from at least two parts of information_DAnd/or A_EAnd/or B_DAnd/or B_EThe value of (c).

Example 37

the mixing mode has at least the following characteristics:

represents said A_DAnd/or A_EThe information of value (B) directly exists in the compressed data code stream and represents the B_DAnd/or B_EThe information of the value of (a) is obtained by means of implicit derivation.

Example 38

the mixing mode has at least the following characteristics:

represents said A_DAnd/or A_EOne part of information of the value (A) directly exists in a compressed data code stream, the other part of information is obtained in an implicit derivation mode, and the A is obtained from at least two parts of information_DAnd/or A_EA value of (B) represents the B_DAnd/or B_EThe information of the value of (a) is obtained by means of implicit derivation.

Example 39

the mixing mode has at least the following characteristics:

represents said A_DAnd/or A_EOne part of information of the value (A) directly exists in a compressed data code stream, the other part of information is obtained in an implicit derivation mode, and the A is obtained from at least two parts of information_DAnd/or A_EThe value of (c).

Example 40

the mixing mode has at least the following characteristics:

is represented by A_XWherein X is D or E, i.e. A_XIs A_DOr A_EInitial value of (iA)_XIs directly present in the compressed data stream, from said initial value iA_XAnd/or at least with said A_XThe values of the A are obtained by calculation and derivation of other coding parameters and/or coding and decoding variables which have no direct relation and/or syntax elements in the compressed data code stream or preset default values of the syntax elements_XThe value of (c).

EXAMPLE 41

the mixing mode has at least the following characteristics:

is represented by A_XWherein X is D or E, i.e. A_XIs A_DOr A_EInitial value of (iA)_XIs directly present in the compressed data stream, from said initial value iA_XAnd/or at least the following coding and decoding variables are calculated, derived and obtained_XThe value of (c):

And/or

Example 42

the mixing mode has at least the following characteristics:

is represented by A_XWherein X is D or E, i.e. A_XIs A_DOr A_EInitial value of (iA)_XIs directly present in the compressed data code stream, said A_XIs equal to iA_X+ dA_XWherein dA_XCalculating, deriving, obtaining from at least the following codec variables:

And/or

Example 43

Embodiment 34 the encoding method or apparatus or the decoding method or apparatus, wherein in the direct mode, the pair a_DAnd A_EPartial combination or all combination of (A)_D, A_E) The values of (1) are subjected to joint binarization, inverse binarization, entropy coding and entropy codingAnd (6) decoding.

Example 44

Embodiment 34 the encoding method or apparatus or the decoding method or apparatus, wherein in the direct mode, the pair a_DThe binarization, inverse binarization, entropy encoding, entropy decoding of the value of (A) is dependent on_EValue or pair A of_EThe binarization, inverse binarization, entropy encoding, entropy decoding of the value of (A) is dependent on_DThe value of (c).

Example 45

Embodiment 34 the encoding method or apparatus or the decoding method or apparatus, wherein in the direct mode, the pair a_DIs zero and A_EThe case of zero is represented by a binary codeword or flag bit (usually 0 or 1) with a code length of one bit.

Example 46

In the encoding method or apparatus, or the decoding method or apparatus, the data is an image or video,

among the sequence parameter sets of the compressed data stream, there are the following syntax elements indicating whether the codec blocks in the corresponding sequence allow the use of the original data component inter prediction:

inter-sps original component prediction allows flag bit sps _ raw _ cross _ component _ prediction _ enabled _ flag

Two-value variable; a value of '1' indicates that the video sequence allows inter-prediction using the original data component; a value of '0' indicates that the video sequence does not use inter-component prediction of the original data;

and/or

In the sequence header of the compressed data stream, the following syntax elements are present to indicate whether the codec blocks in the corresponding sequence allow the use of the original data inter-component prediction:

sequence original inter-component prediction allows flag bit seq _ raw _ cross _ component _ prediction _ enabled \ flag

And (4) binary variables. A value of '1' indicates that the video sequence allows inter-prediction using the original data component; a value of '0' indicates that the video sequence does not use inter-component prediction of the original data;

and/or

Among the picture parameter sets of the compressed data stream, there are the following syntax elements indicating whether the codec block in the corresponding picture allows the use of the original data component inter prediction:

pps original inter-component prediction enabled flag bit pps _ raw _ cross _ component _ prediction _ enabled \ flag

Two-value variable; a value of '1' indicates that the image allows inter-prediction using the original data components; a value of '0' indicates that the image does not use inter-raw data component prediction;

and/or

In the picture header of the compressed data stream, the following syntax elements are present to indicate whether the codec block in the corresponding picture allows the use of the original data component inter prediction:

inter-picture original component prediction enabled flag bit pic _ raw _ cross _ component _ prediction _ enabled \u flag

Two-value variable; a value of '1' indicates that the image allows inter-prediction using the original data components; a value of '0' indicates that the image does not use inter-raw data component prediction.

And/or

In the slice header of the compressed data code stream, the following syntax elements are present to indicate whether the codec blocks in the corresponding slice allow the use of the original data component inter prediction:

inter-slice raw component prediction allows flag bit slice _ raw _ cross _ component _ prediction \ u enabled_flag

Two-value variable; a value of '1' indicates that the slice allows inter-prediction using the original data component; a value of '0' indicates that the slice does not use inter-raw data component prediction.

And/or

Among the sequence parameter sets of the compressed data stream, there are the following syntax elements indicating whether the coded blocks in the corresponding sequence allow the use of residual data component inter prediction:

inter-sps residual component prediction allows flag bit sps _ residual _ cross _ component _ prediction _ enabled_flag

Two-value variable; a value of '1' indicates that the video sequence allows inter-component prediction using residual data; a value of '0' indicates that the video sequence does not use residual data inter-component prediction;

and/or

In the sequence header of the compressed data stream, the following syntax elements are present to indicate whether the coded blocks in the corresponding sequence allow the use of residual data inter-component prediction:

sequence residual inter-component prediction enabled flag bit seq _ residual _ cross _ component _ prediction _ enabled_flag

And (4) binary variables. A value of '1' indicates that the video sequence allows inter-component prediction using residual data; a value of '0' indicates that the video sequence does not use residual data inter-component prediction;

and/or

Among the picture parameter sets of the compressed data stream, there are the following syntax elements indicating whether the coded blocks in the corresponding picture allow the use of residual data component inter prediction:

pps residual inter-component prediction enable flag bit pps _ residual _ cross _ component _ prediction _ enabled_flag

Two-value variable; a value of '1' indicates that the picture allows inter-component prediction using residual data; a value of '0' indicates that the picture does not use residual data inter-component prediction;

and/or

In the picture header of the compressed data stream, the following syntax elements are present to indicate whether the coded and decoded blocks in the corresponding picture allow the use of residual data inter-component prediction:

inter-picture residual component prediction enabled flag bit pic _ residual _ cross _ component _ prediction _ enabled_flag

and/or

In the slice header of the compressed data stream, the following syntax elements exist to indicate whether the coded blocks in the corresponding slice allow the use of residual data inter-component prediction:

inter-slice residual component prediction enabled flag bit slice _ residual _ cross _ component _ prediction _ enabled_flag

Two-value variable; a value of '1' indicates that the slice allows inter-component prediction using residual data; a value of '0' indicates that the slice does not use residual data inter-component prediction.

Example 47

In the encoding method or apparatus or the decoding method or apparatus, when encoding and decoding are performed using at least the original data component inter-prediction and/or the residual data component inter-prediction and a prediction difference value of one component generated is subjected to quantization and inverse quantization operations, a quantization degree of the component is adjusted; one way of adjusting is to increase or decrease the quantization parameter also to the value of the quantization factor QP, where the amount of increase or decrease is a predetermined constant, typically less than 10.

Embodiment 48 (version on "two-component prediction" corresponding to embodiment 13 or 28 on "three-component prediction")

In the encoding method or device or the decoding method or device, original data and/or residual data are simply called original/residual data, the original/residual data have at least two components D and E, and the E and the D are used for predicting the D and the E respectively;

in the encoding method or apparatus, the operation of constructing the prediction value predC = (predD, predE) and the prediction difference value Δ C = (Δ D, Δ E) of the original/residual data from the original/residual data E and D of at least the two components includes at least the operation expressed by the following formula:

predC = (-a*E, a*D)

and

(∆D, ∆E) = ∆C = (C－PredC)/d = ((D, E)－(-a*E, a*D))/d = ((D + a*E,E－a*D))/d

where a is a multiplication factor, d =1 or d =1 + a²(ii) a Obviously, Δ D = (a D + a)²*E)/d，a*∆E =(a*E－a²D/D, thereby deriving D-a E = (D + a E- (a E-a)²*D))/d = (D + a²Δ D/D and Δ E + a = (E-a. D + a)²*E)/d = (E + a²E)/D, it can be seen that the predicted difference D and E is the relationship between the two components D and E:

D = (∆D－a*∆E)*d/(1 + a²) And E = (E + a) Δ D/(1 + a)²)；

In the decoding method or device, the calculation modes of constructing the original/residual data reconstruction values recD and recE of the two components from the prediction difference reconstruction values recΔ D and recE of at least the two components are as follows:

recD = (rec∆D－a*rec∆E)*d/(1 + a²) And recE = (recE + a rec. D)/(1 + a)²)，

Wherein, the prediction difference reconstruction values rec Δ D and rec Δ E are the results of the coding operations including prediction and/or transformation and/or quantization and the inverse operation thereof on the prediction differences D and E; obviously, when a =0, the predicted value predC of the raw/residual data is zero, the raw/residual data is equal to the prediction difference value, and the raw/residual data reconstruction value is equal to the prediction difference value reconstruction value.

Embodiment 49 (version on "two-component prediction" corresponding to embodiment 15 or 30 on "three-component prediction")

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 48, a value range of the multiplication factor a is one of the following sets:

R = {

: -2ⁿ⁺¹≤ m ≤ 2ⁿ⁺¹}={0, ±

: 1 ≤ m ≤ 2ⁿ⁺¹} = {±

Or

A subset of R, comprising at least:

R_n= {0,

: m=±2^k, m=±(2ⁿ+2^k) K is 0. ltoreq. n or R_nIncluding the subset 0,

: m=±2^k0 ≦ k ≦ n } or a subset thereof

Or

When n =6, the number of the bits is increased,

, ±

,±

, ±

, ±

, ±

, ±1, ±1

, ±1

, ±1

, ±1

, ±1

, ±1

, + -2 } or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

, ±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

, ±

, ±1}={0,±

, ±

, ±

, ±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±1}={0,±

, ±

, + -1} or a subset thereof

Or

R₆A subset of {0, +/-

, ±

, ±

, ±

}={0, ±

, ±

, ±

, ±

Or a subset thereof

Or

When n =1, the number of the bits is set to n =1,

R₁= {0,

: m=±2^k, m=±(2+2^k), 0 ≤ k ≤ 1} = {0,

m = ± 1, ± 2, ± 3, ± 4} or a subset thereof

Or

R₁A subset of {0, +/-

, + -1, + -2 } or a subset thereof

Or

R₁A subset of {0, +/-

, + -1} or a subset thereof

Or

R₁Or a subset thereof {0, 1} or {0, -1 }.

Example 50

Embodiment 48 the encoding method or apparatus or the decoding method or apparatus, wherein d =1 + a²，

When a is not zero, the recE or rec D is set to zero, and in the decoding method or device, the original/residual data reconstruction values recD and recE of the two components are constructed from at least rec D or rec E in the following calculation mode:

recD = recΔ D and recE = a rec D

Or

recD = -a recΔ E and recE = rec E;

when a is zero, the original/residual data reconstruction values recD and recE of the two components are calculated in the following way:

recD = recΔ D and recE = recE.

Example 51

The encoding method or apparatus or the decoding method or apparatus according to embodiment 50,

when a is not zero, the compressed data code stream at least has the information needed for representing and obtaining the reconstruction value rec, but does not have the information needed for representing and obtaining the reconstruction value rec;

when a is zero, the compressed data code stream at least has the information needed to represent and obtain the reconstruction values of the prediction difference values recΔ D and rec Δ E of the two components.

Example 52

The example 51 is the encoding method or apparatus or the decoding method or apparatus, where the information indicating and obtaining the Δ D and/or the Δ E includes at least information indicating and obtaining the result of the encoding operation including the prediction and/or the transformation and/or the quantization of the prediction difference Δ D and/or Δ E or at least information indicating and obtaining the result of the encoding operation including the quantization of the prediction difference Δ D and/or Δ E.

Example 53

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 50, a range of values of the multiplicative factor a is {0, +/-

, + -1, + -2 } or a subset thereof.

Example 54

Embodiment 53 as described in the encoding method or apparatus or the decoding method or apparatus, wherein when the intra coding mode is adopted for the encoding/decoding block, the range of the multiplication factor a is R₁₁={0,

1 or R₁₂={0, -

-1}, otherwise, the value range of the multiplication factor a is R₂₁=0, 1 or R₂₂={0, -1}。

Example 55

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 54, in a picture parameter set or a picture header or a slice header of a compressed data code stream, the following syntax elements exist to indicate a value range of an inter-component prediction multiplication factor a in all coding and decoding blocks in a corresponding picture or slice:

value range flag bit pps _ ccp _ multiplier _ value _ range _ flag of pps inter-component prediction multiplication factor

Or

Flag bit pic _ ccp _ multiplier _ value _ range _ flag of multiplication factor value range for image component prediction

Or

Band component inter-prediction multiplication factor value range flag bit slice _ ccp _ multiplier _ value _ range _ flag

Two-value variable; the value of '1' indicates that in the corresponding image or strip, when the encoding and decoding block adopts the intra-frame encoding mode, the value range of the multiplication factor a is R₁₁Otherwise, the value range of the multiplication factor a is R₂₁(ii) a The value of '0' indicates that in the corresponding image or strip, when the encoding and decoding block adopts the intra-frame encoding mode, the value range of the multiplication factor a is R₁₂Otherwise, the value range of the multiplication factor a is R₂₂。

Example 56

In the encoding method or apparatus or the decoding method or apparatus according to embodiment 50, the following syntax elements whose values are binary variables exist in a header of a coding/decoding block of a compressed data code stream:

coded block component D mode flag bit cb _ D _ md _ flag，Coded block component E mode flag bit cb _ E _ md _ flag，Weaving machine Code block component zero flag bit cb _ be _0_ flagThe values of the three flag bits determine the value of a and which component is set to zero,

when the coding block adopts intra coding mode, the zeroed components and the value of a are determined by the following table:

when the codec block does not employ intra coding mode, the zeroed components and the value of a are determined by the following table:

cb_D_md_flag	cb_E_md_flag	cb_be_0_flag	component of zero setting	Value of a
					x	x	0	Is free of	a=0
1	1	1	rec is set to zero	a =1 or-1

In the above table, whether the value of a, which is not zero, is positive or negativeValue range flag bit for multiplication factor predicted between pps componentsOrValue range flag bit of prediction multiplication factor between image componentsOrBand component inter-prediction multiplication factor value range flag bitThe value of x is determined, wherein the value of x is 0 or 1, the value of 0 indicates that the prediction difference values of the corresponding components D or E in the coding and decoding block are all zero, and the value of 1 indicates that the prediction difference values of the corresponding components D or E in the coding and decoding block are not all zero.

Example 57

The encoding method or apparatus or the decoding method or apparatus as in embodiment 56, wherein the right shift by a =1/2 is 1 bit, and the right shift by a = -1/2 is 1 bit after the multiplication by-1.

Claims

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

2. A decoding method or apparatus for compressing multi-component data, comprising at least the steps or modules of:

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;

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 pixel components.

4. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, wherein the data is an image or video having three components, one primary component and two secondary components.

5. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, wherein the data is an image or video having three components, one primary component and two secondary components, the three components being one of:

a Y luminance component, a U chrominance component, a V chrominance component,

a Y luminance component, a Cb chrominance component, a Cr chrominance component,

r red component, G green component, B blue component,

a G green component, an R red component, a B blue component,

g green component, B blue component, R red component.

6. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, wherein the data is an image or video having three components, one primary component and two secondary components, original data and/or residual data are simply referred to as original/residual data, and in the original/residual data inter-component prediction,

And/or

7. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, wherein the data is an image or video of 422-sample format having three components, one primary component and two secondary components, the sampling rate and size of the two secondary components being one-half of the sampling rate and size of the primary component, respectively, the two secondary components are predicted using the primary component in the original data inter-component prediction and the residual data inter-component prediction, and the operation of the original data inter-component prediction and the operation of the residual data inter-component prediction each include an operation of horizontal 2:1 down-sampling or vertical 2:1 down-sampling the primary component.

8. The encoding method or apparatus according to claim 1 or the decoding method or apparatus according to claim 2, wherein the data is an image or video of a 420-sample format having three components, one primary component and two secondary components, the sampling rate and size of the two secondary components being one quarter of the sampling rate and size of the primary component, respectively, the two secondary components are predicted using the primary component in the original data inter-component prediction and the residual data inter-component prediction, and the operations of the original data inter-component prediction and the residual data inter-component prediction each include operations of horizontal 2:1 down-sampling and vertical 2:1 down-sampling total 4:1 down-sampling of the primary component.

9. The encoding method or apparatus or the decoding method or apparatus according to claim 8, wherein the 4:1 downsampling is to calculate a weighted average value of I × J principal components as a value of a downsampled principal component according to one of the following downsampling formulas:

10. The encoding method or apparatus or the decoding method or apparatus according to claim 8, wherein information specifying a downsampling operation on the principal component by selecting one of predetermined downsampling formulas exists in a compressed data code stream.