WO2014166433A1

WO2014166433A1 - Depth-image encoding and decoding methods and encoding/decoding device

Info

Publication number: WO2014166433A1
Application number: PCT/CN2014/075234
Authority: WO
Inventors: 郑萧桢
Original assignee: 华为技术有限公司
Priority date: 2013-04-12
Filing date: 2014-04-14
Publication date: 2014-10-16
Also published as: CN103200406A; CN103200406B

Abstract

Embodiments of the present invention provide depth-image encoding and decoding methods and an encoding/decoding device. The encoding method comprises: obtaining predicted data corresponding to a current image block of a depth-image; in accordance with the preset step size, obtaining a predicted pixel value from within the predicted data; calculating, in accordance with the predicted pixel value, a first average value for the predicted data; the preset step size is a positive integer other than 1; obtaining, in accordance with the first average value for the predicted data and with a pixel value for the pixels of the current image block, residuals of the current image block; encoding the residuals of the current image block; encoding and decoding efficiency can thus be improved.

Description

Decoding method and codec device for depth image

TECHNICAL FIELD Embodiments of the present invention relate to image processing technologies, and in particular, to a codec method and a codec device for a depth image. Background Art In a three-dimensional video image system, two types of images, a visual image and a depth image, are included, and the visual image contains information of the image itself, such as color information, and the depth image contains depth information of the visual image, such as contour information. In the transmission of the three-dimensional video image, in order to reduce the bandwidth occupied by the video transmission, the video image needs to be encoded, including encoding the visual image and the depth image to reduce the size of the three-dimensional video image data.

In the prior art, an intra-frame prediction method is used for encoding a depth image in a three-dimensional video image system. First, the encoding end divides the image block to be encoded into one or more sub-image blocks, obtains an intra prediction mode of the current image block, and obtains prediction data of the current image block according to the intra prediction mode, where the image block and the image block The divided sub-image blocks are generally referred to collectively as image blocks, and the prediction data of the image blocks is used as prediction data of the current image block codec, which is derived from pixel data of adjacent blocks that have been encoded or decoded, or codec. A fixed value preset by the system, or other predictive data from the neighboring block for the current block codec. The encoding end performs a point-by-point difference between the pixel value of each pixel in the image block and the pixel value of each pixel corresponding to the prediction data (or the data of the reference block generated by the prediction data) to obtain a residual matrix, and the residual matrix The difference matrix performs quantization, transform and entropy coding, and sends the code stream obtained by entropy coding to the decoding end. The decoding end reads the bit stream from the obtained code stream, performs inverse quantization and inverse coding, and obtains a corresponding residual, obtains prediction data of the current image block according to the intra prediction mode, and uses the residual of the current image block and the prediction data. The pixel values (or the pixel values of the reference blocks generated from the prediction data) are added to obtain the pixel values of the current image block.

However, when the size of the image block is large, the pixel average of the image block and the pixel mean of the reference image block are calculated by the prior art method, and the encoding and decoding efficiency is not high.

Summary of the invention Embodiments of the present invention provide a codec method and a codec device for a depth image to improve codec efficiency. A first aspect of the embodiments of the present invention provides a method for encoding a depth image, including: obtaining prediction data of a current image block of a depth image, and obtaining a predicted pixel value in the prediction data according to a preset step size, according to the prediction The pixel value calculates a first mean value of the prediction data, and the preset step size is a positive integer that is not 1;

Obtaining a residual of the current image block according to the first mean value of the prediction data and the pixel value of the pixel point of the current image block;

The residual of the current image block is encoded.

With reference to the first aspect, in a first possible implementation, the obtaining a residual according to a first average of the prediction data and a pixel value of a pixel of the image block includes:

Performing a difference between a pixel point of the current image block and a first mean value of the prediction data of the current image block to obtain a residual of the current image block; or

Obtaining data of the current image block of the depth image, obtaining a predicted pixel value in the data of the current image block according to a preset step size, and calculating data of the current image block according to the predicted pixel value of the data of the current image block. a second mean, a difference between the second mean value of the data of the current image block and the first mean value of the prediction data of the current image block, to obtain a residual of the current image block.

With reference to the first aspect, in a second possible implementation, the obtaining the predicted pixel value in the prediction data according to the preset step size includes:

Obtaining a predicted pixel value by performing equal steps in the horizontal direction and the vertical direction according to the size of the current image block; or

Obtaining a predicted pixel value by performing unequal steps in the horizontal direction and the vertical direction according to the size of the current image block; or

Determining the preset step size according to the size of the current image block, and obtaining a predicted pixel value according to the preset step size.

With reference to the first aspect, or any one of the first to the second possible implementations of the first aspect, in a third possible implementation, the method further includes:

The value of the preset step size is encoded.

With reference to the first aspect, or any one of the first to the third possible implementation manners of the first aspect, in a fourth possible implementation, the preset step size is used to identify the prediction The interval between data position coordinates.

With reference to the first aspect, or any one of the first to the fourth possible implementation manners of the first aspect, in a fifth possible implementation, the residual of the current image block Coding, including:

Mapping the residual to a residual mapping value, indicating that the bit of the residual mapping value is smaller than a bit representing the residual;

The residual map value is encoded.

With reference to the first aspect, or any one of the possible implementation manners of the first to fifth possible implementation manners of the first aspect, in a sixth possible implementation manner, Before obtaining predicted pixel values in the forecast data, it also includes:

Determining, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset if, if it is, determining to obtain the predicted pixel value in the prediction data according to the preset step size.

With reference to the sixth possible implementation manner, in a seventh possible implementation manner, the intra prediction mode set includes at least one intra prediction mode:

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

A second aspect of the embodiments of the present invention provides a method for decoding a depth image, including: obtaining prediction data of a current image block of a depth image, and obtaining a predicted pixel value in the prediction data according to a preset step size, according to the prediction The pixel value calculates a first mean value of the prediction data, and the preset step size is a positive integer that is not 1;

Obtaining a residual of the current image block;

A pixel value of a pixel of the current image block is obtained according to a first mean of the prediction data and a residual of the current image block.

With reference to the second aspect, in a first possible implementation, the obtaining the pixel value of the pixel of the current image block according to the first average of the prediction data and the residual of the current image block includes: Adding a first mean value of the prediction data of the current image block to a residual of the current image block to obtain a pixel value of a pixel point of the current image block; or

Adding a first mean value of the prediction data of the current image block to the residual, obtaining a pixel mean of the current image block, and using a pixel mean of the current image block as a pixel point of the current image block. Pixel values.

With reference to the second aspect, in a second possible implementation, the obtaining the predicted pixel value in the prediction data according to the preset step size includes:

Determining the preset step size according to the size of the current image block, and obtaining the predicted pixel value according to the preset step size.

With reference to the second aspect, or any one of the first to the second possible implementation manners of the second aspect, in a third possible implementation manner, the method further includes:

The parsing code stream obtains a value identifying the step size.

With reference to the second aspect, or any one of the first to the third possible implementation manners of the second aspect, in a fourth possible implementation, the preset step size is used to identify Predict the interval between data position coordinates.

With reference to the second aspect, or any one of the first to the fourth possible implementation manners of the second aspect, in a fifth possible implementation manner, the obtaining the residual of the current image block Poor, including:

Performing a decoding process on the encoding result of the current image block to obtain a residual mapping value of the current image block;

Mapping the residual mapping value to a residual, using the residual as a residual of the current image block, indicating that the bit of the residual mapping value is smaller than a bit representing the residual.

With reference to the second aspect, or any one of the possible implementation manners of the first to fifth possible implementation manners of the second aspect, in the sixth possible implementation manner, Before obtaining the predicted pixel value in the predicted data, the method further includes:

Determining whether the intra prediction mode belongs to a preset according to an intra prediction mode of a current image block If so, it is determined that the predicted pixel value is obtained in the prediction data according to the preset step size.

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

A third aspect of the embodiments of the present invention provides a coding apparatus for a depth image, including: a first processing module, configured to obtain prediction data of a current image block of a depth image, and obtain a prediction in the prediction data according to a preset step size. a pixel value, a first average value of the prediction data is calculated according to the predicted pixel value, where the preset step size is a positive integer that is not 1;

a second processing module, configured to obtain a residual of the current image block according to the first mean value of the prediction data and the pixel value of the pixel point of the current image block;

And an encoding module, configured to encode a residual of the current image block.

With reference to the third aspect, in a first possible implementation, the second processing module is configured to perform a difference between a pixel point of the current image block and a first average value of prediction data of the current image block, a residual of the current image block; or obtaining data of a current image block of the depth image, obtaining a pixel value in the data of the current image block according to a preset step size, according to a pixel value of the data of the current image block Calculating a second mean value of the data of the current image block, and comparing a second mean value of the data of the current image block with a first mean value of the predicted data of the current image block to obtain a residual of the current image block .

With reference to the third aspect, in a second possible implementation manner, the first processing module is configured to obtain a predicted pixel value according to the equal size of the horizontal direction and the vertical direction according to the size of the current image block; or Obtaining a predicted pixel value according to the unequal step size of the horizontal direction and the vertical direction according to the size of the current image block; or determining the preset step size according to the size of the current image block, according to the preset The step size gets the predicted pixel value.

With reference to the third aspect, or any one of the possible implementations of the first to the second possible implementations of the third aspect, in a third possible implementation, the coding module is further configured to The value of the preset step size is encoded.

With reference to the third aspect, or any one of the first to the third possible implementation manners of the third aspect, in a fourth possible implementation, the preset step size is used to identify the prediction The interval between data position coordinates.

With reference to the third aspect, or any one of the first to the fourth possible implementation manners of the third aspect, in a fifth possible implementation, the coding module is specifically used to The residual is mapped to a residual map value, and the bit of the residual map value is smaller than a bit representing the residual, and the residual map value is encoded.

With reference to the third aspect, or any one of the first to the fifth possible implementation manners of the third aspect, in a sixth possible implementation, the first processing module is further configured to Determining, by the intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set, and if so, determining to obtain the predicted pixel in the prediction data according to the preset step size value.

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

A fourth aspect of the embodiments of the present invention provides a decoding apparatus for a depth image, including: a processing module, configured to obtain prediction data of a current image block of a depth image, and obtain a predicted pixel value in the prediction data according to a preset step size. Calculating a first mean value of the prediction data according to the predicted pixel value, where the preset step size is a positive integer that is not 1;

And a decoding module, configured to obtain a residual of the current image block, and obtain a pixel value of a pixel of the current image block according to a first average of the prediction data and a residual of the current image block.

With reference to the fourth aspect, in a first possible implementation, the decoding module is specifically configured to add a first average value of the prediction data of the current image block to a residual of the current image block, to obtain the a pixel value of a pixel of the current image block; or, adding a first mean value of the prediction data of the current image block to the residual to obtain a pixel mean of the current image block, where the current image block is The pixel mean is the pixel value of the pixel of the current image block. With reference to the fourth aspect, in a second possible implementation, the processing module is specifically configured to obtain a predicted pixel value according to an equal step size of a horizontal direction and a vertical direction according to a size of the current image block; or Determining the size of the current image block to obtain a predicted pixel value in an unequal step in the horizontal direction and the vertical direction; or determining the preset step size according to the size of the current image block, according to the preset step size The predicted pixel value is obtained.

With reference to the fourth aspect, or any one of the possible implementations of the first to the second possible implementations of the fourth aspect, in a third possible implementation, the decoding module is further configured to obtain a code stream. Identifies the value of the step size.

With reference to the fourth aspect, or any one of the first to the third possible implementation manners of the fourth aspect, in a fourth possible implementation, the preset step size is used to identify the prediction The interval between data position coordinates.

With reference to the fourth aspect, or any one of the first to the fourth possible implementation manners of the fourth aspect, in a fifth possible implementation, the decoding module is specifically configured to use the current Performing a decoding process on the coding result of the image block to obtain a residual mapping value of the current image block, mapping the residual mapping value into a residual, and using the residual as a residual of the current image block, indicating The bit of the residual map value is less than the bit representing the residual.

With reference to the fourth aspect, or any one of the first to the fifth possible implementation manners of the fourth aspect, in a sixth possible implementation, the processing module is further configured to use the current image block The intra prediction mode determines whether the intra prediction mode belongs to a preset intra prediction mode set, and if so, determines to obtain the predicted pixel value in the prediction data according to the preset step size.

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

A fifth aspect of the embodiments of the present invention provides a method for encoding a depth image, including: obtaining prediction data of a current image block of a depth image, performing arithmetic operations on each N pixel points in the pixel in the prediction data, and obtaining a corresponding pixel value. Arithmetic value, arithmetic according to the corresponding pixel value Calculating a first mean value of the predicted data;

The residual of the current image block is encoded.

With reference to the fifth aspect, in a first possible implementation manner, performing an arithmetic operation on each of the pixel points in the pixel in the prediction data to obtain an arithmetic value of the corresponding pixel value, including:

An averaging operation is performed on every N pixel points in the pixel in the prediction data to obtain an average pixel value, and the average pixel value is taken as an arithmetic value of the corresponding pixel.

With reference to the fifth aspect, or the first possible implementation manner of the fifth aspect, in a second possible implementation, the residual is obtained according to the first average of the prediction data and the pixel value of the pixel of the image block , including:

Performing a difference between a pixel point of the image block and a first mean value of prediction data of the current image block to obtain a residual of the current image block;

Or obtaining data of the current image block of the depth image, obtaining the pixel value in the data of the current image block according to a preset step size, and calculating a second mean value of the data of the image block according to the pixel value, The second mean of the image block is different from the first mean of the reference image block of the current image block to obtain a residual of the current image block.

A sixth aspect of the embodiments of the present invention provides a method for decoding a depth image, including: obtaining prediction data corresponding to a current image block of a depth image, performing arithmetic operations on each N pixel points in the pixel in the prediction data, and obtaining corresponding pixels. An arithmetic value of the value, calculating a first mean value of the prediction data according to an arithmetic value of the corresponding pixel value;

Obtaining a residual of the current image block;

A pixel value of a pixel point of the current image block is obtained according to an average of the prediction data and a residual of the current image block.

With reference to the sixth aspect, in a first possible implementation manner, performing an arithmetic operation on each of the pixel points in the pixel in the prediction data to obtain an arithmetic value of the corresponding pixel value, including:

And averaging every N pixels in the current image block and the pixel in the prediction data to obtain an average pixel value, and using the average pixel value as an arithmetic value of the corresponding pixel value.

With reference to the sixth aspect, or the first possible implementation manner of the sixth aspect, in a second possible implementation, the current is obtained according to the mean value of the prediction data and a residual of the current image block The pixel value of the pixel of the image block, including:

Adding a mean value of the prediction data of the current image block to a residual of the current image block, and obtaining a pixel value of a pixel point of the current image block;

Or, adding the mean value of the prediction data of the current image block to the residual, obtaining a pixel mean of the current image block, and using a pixel average of the current image block as a pixel point of the current image block. Pixel values.

A seventh aspect of the embodiments of the present invention provides a coding apparatus for a depth image, including: a first processing module, configured to obtain prediction data corresponding to a current image block of a depth image, and for each N pixel points in a pixel in the prediction data Performing an arithmetic operation, obtaining an arithmetic value of a corresponding pixel value, and calculating a first mean value of the predicted data according to an arithmetic value of the corresponding pixel value;

With reference to the seventh aspect, in a first possible implementation manner, the first processing module is specifically configured to perform an averaging operation on each of the N pixel points in the predicted data to obtain an average pixel value, The average pixel value is taken as the arithmetic value of the corresponding pixel.

With reference to the seventh aspect, or the first possible implementation manner of the seventh aspect, in a second possible implementation, the second processing module is specifically configured to: the pixel of the image block and the current image block Obtaining a difference between the first mean value of the prediction data to obtain a residual of the current image block; or obtaining data of a current image block of the depth image, obtaining the pixel in data of the current image block according to a preset step size a value, a second mean value of the data of the image block is calculated according to the pixel value, and a second mean value of the image block is compared with a first mean value of a reference image block of the current image block to obtain the current image. The residual of the block.

An eighth aspect of the present invention provides a decoding apparatus for a depth image, including:

a processing module, configured to obtain prediction data corresponding to a current image block of the depth image, perform an arithmetic operation on each N pixel points in the pixel in the prediction data, and obtain an arithmetic value of the corresponding pixel value, according to an arithmetic value of the corresponding pixel value Calculating a first mean value of the predicted data;

And a decoding module, configured to obtain a residual of the current image block, and obtain a pixel value of a pixel of the current image block according to an average of the prediction data and a residual of the current image block.

With reference to the eighth aspect, in a first possible implementation, the processing module is specifically configured to An average operation is performed for every N pixels in the current image block and the pixel in the prediction data to obtain an average pixel value, and the average pixel value is used as an arithmetic value of the corresponding pixel value.

With reference to the eighth aspect, or the first possible implementation manner of the eighth aspect, in a second possible implementation, the decoding module is specifically configured to use an average of prediction data of the current image block and the current image Adding a residual of the block to obtain a pixel value of a pixel of the current image block; or, adding a mean value of the prediction data of the current image block to the residual to obtain a pixel mean of the current image block And using a pixel average of the current image block as a pixel value of a pixel point of the current image block.

The codec method and the codec device of the depth image provided by the embodiment of the present invention obtain the predicted pixel value in the prediction data according to the preset step size by obtaining the prediction data corresponding to the current image block of the depth image, according to the prediction. a pixel value calculates a first mean value of the prediction data, the preset step size is a positive integer that is not 1; obtaining a current image according to a first mean value of the prediction data and a pixel value of a pixel point of the current image block Residual of the block; encoding the residual of the current image block, may reduce the number of reference pixels of the prediction data of the current image block, and thus, when calculating the first mean value of the prediction data of the current image block, The amount of calculation is significantly reduced, so that the coding efficiency can be improved. Correspondingly, when the decoding end performs the decoding process on the current image block, the calculation amount of the decoding process using the decoding method corresponding to the encoding method of the encoding end is also significantly reduced. Therefore, the technical solution of the embodiment of the present invention can be effectively improved. Codec efficiency.

BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief description of the drawings used in the embodiments or the prior art description will be briefly described below. The drawings are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative work.

1 is a schematic flow chart of Embodiment 1 of a method for encoding a depth image according to the present invention;

2 is a schematic diagram of an image block according to Embodiment 1 of a method for encoding a depth image according to the present invention;

3 is a schematic flowchart of Embodiment 2 of a method for encoding a depth image according to the present invention;

4 is a schematic flowchart of Embodiment 1 of a method for decoding a depth image according to the present invention; 5 is a schematic flowchart of Embodiment 2 of a method for decoding a depth image according to the present invention; FIG. 6 is a schematic structural diagram of Embodiment 1 of a device for encoding a depth image according to the present invention;

7 is a schematic structural diagram of Embodiment 1 of a decoding apparatus for a depth image according to the present invention;

8 is a schematic flowchart of Embodiment 3 of a method for encoding a depth image according to the present invention;

9 is a schematic flowchart of Embodiment 3 of a method for decoding a depth image according to the present invention;

10 is a schematic structural diagram of Embodiment 2 of a depth image encoding apparatus according to the present invention;

FIG. 11 is a schematic structural diagram of Embodiment 2 of a decoding apparatus for a depth image according to the present invention. The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. The embodiments are a part of the embodiments of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

In a 3D video system, the depth image contains depth information of the view image, for example, contour information. The depth image is characterized by a smoother texture in most areas. Usually, only the edge information of Rayleigh is included at the boundary of the object, and the pixels in the depth image are The values are relatively simple and the distribution is uniform. Therefore, when each prediction of the depth image is encoded by the intra prediction method, the pixel mean of the image block to be encoded is used between the pixel mean of the reference image block corresponding to the image block. The residual is coded. When encoding a large-size image block, it is necessary to calculate the mean of the image block or the average of the reference image block. When the depth image block is divided into image blocks, the size is usually larger. The image block contains less texture information, and the embodiment of the present invention combines the feature that the large size image block contains less texture information, between the pixel mean value of the image block to be encoded and the pixel mean value of the reference image block corresponding to the image block. Before the residual is encoded, the image block is identical to the reference image block corresponding to the image block. According to the punctured pixel, the pixel mean of the reference image block and the pixel mean of the reference image block corresponding to the image block are obtained, so as to reduce the calculation amount at the time of encoding and improve the coding efficiency. The technical solutions of the present invention are described in detail below in several specific embodiments.

1 is a schematic flowchart of a first embodiment of a method for encoding a depth image according to the present invention. The execution body of the embodiment is an encoding device, and the method of this embodiment includes:

S101: Obtain a prediction data corresponding to a current image block of the depth image, according to a preset step size The predicted pixel value is obtained in the measured data, and the first mean value of the predicted data is calculated according to the predicted pixel value, and the preset step size is a positive integer not being 1.

In this embodiment, the preset step size may be used to identify an interval between predicted data position coordinates.

S102. Obtain a residual according to the first mean of the prediction data and the pixel value of the pixel of the current image block.

S103. Encode a residual of the current image block.

Specifically, the process of obtaining the predicted pixel value in the prediction data according to the preset step size by the encoding end is a process of performing the sampling on the prediction data.

2 is a schematic diagram of an image block of a first embodiment of a method for encoding a depth image according to the present invention. As shown in FIG. 2, each square represents a pixel, and the size of the prediction data of the current image block is 8 x 8, that is, 64 pixels. pixel. With a preset step size of 2, the predicted data of the current image block can be processed 2:1, that is, one pixel is spaced once in the horizontal direction and the vertical direction, and one pixel is obtained. The predicted pixel value can be obtained. The black pixel shown in Fig. 2 is the predicted pixel value, and the number of predicted pixel values is 16.

Then, the encoding end may calculate a first mean value of the prediction data according to the predicted pixel value, and obtain a residual according to the first mean value of the predicted data and the pixel value of the pixel point of the image block.

Correspondingly, when the decoding process is performed on the current image block by the decoding end, the reference image block corresponding to the current image block may be processed by the processing procedure corresponding to the encoding end, thereby further improving the decoding efficiency.

In this embodiment, by obtaining the prediction data corresponding to the current image block of the depth image, obtaining the predicted pixel value in the prediction data according to the preset step size, and calculating the first mean value of the prediction data according to the predicted pixel value, the preset step size is a positive integer not being 1; a residual is obtained according to the first mean of the prediction data and the pixel value of the pixel of the current image block; encoding the residual of the current image block, the reference pixel of the prediction data of the current image block may be reduced Therefore, when calculating the first mean value of the prediction data of the current image block, the calculation amount is significantly reduced, so that the coding efficiency can be improved. Correspondingly, when the decoding end performs decoding processing on the current image block, the calculation amount of the decoding process using the decoding method corresponding to the encoding method of the encoding end is also significantly reduced. Therefore, the technical solution of the embodiment can effectively improve the editing. Decoding efficiency.

In the foregoing embodiment, when S102 is specifically implemented, for example, the following two manners can be implemented: Method 1: The pixel of the current image block is compared with the first average of the prediction data of the current image block. Poor, get the residual of the current image block; or,

The second method is: obtaining data of the current image block of the depth image, obtaining a predicted pixel value in the data of the current image block according to the preset step size, and calculating a second mean value of the data of the current image block according to the pixel value of the data of the current image block, The second mean value of the data of the current image block is compared with the first mean value of the prediction data of the current image block to obtain a residual of the current image block, and at this time, the image block has only one residual value.

Compared with the first method, the foregoing method 2 can reduce the number of reference pixel points of the data of the current image block, so that when calculating the second mean value of the data of the current image block, the calculation amount is significantly reduced, thereby further improving the coding efficiency. .

It should be noted that, in the second method, the process of obtaining the pixel value in the data of the current image block according to the preset step size is similar to the process of obtaining the predicted pixel value in the prediction data of the current image block according to the preset step size. , will not repeat them here.

The following is a detailed description of the technical solution of the present invention. FIG. 3 is a schematic flowchart of the second embodiment of the method for encoding a depth image according to the present invention. The execution subject of the embodiment is an encoding device, and the method of the embodiment includes :

S301: Determine, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set; if yes, execute S302~306, if not, execute S307.

S302: Determine to perform a sample processing on the current image block and the reference image block corresponding to the current image block.

The intra prediction mode set in this embodiment includes at least a DC mode, a Planar mode, a displayed identification mode, and an intra prediction mode based mode, and the current image block and the reference image block are performed according to a preset lowering rule. Before the sample processing, it is determined whether the intra prediction mode belongs to the intra prediction mode set, and correspondingly, the decoding end has the same intra prediction mode set as the encoding end, before the reference image block is sampled, It is also necessary to first determine whether the intra prediction mode belongs to a set of intra prediction modes. The preset set of intra prediction modes is a set of intra prediction modes that have been tested to show that the encoding method of S302 S306 can obtain better encoding processing performance.

If it belongs, it is determined that the current image block and the reference image block corresponding to the current image block are subjected to the sample processing.

It should be noted that steps S301 and S302 are optional steps, and may not be performed, and S303~S306 are directly executed. S303: Obtain data of a current image block of the depth image, obtain a predicted pixel value in the data of the current image block according to the preset step size, obtain prediction data corresponding to the current image block of the depth image, and obtain the predicted data according to the preset step size. Get the predicted pixel value.

Optionally, obtaining the predicted pixel value in the data of the current image block according to the preset step size, or obtaining the predicted pixel value in the prediction data according to the preset step size may be implemented in the following three manners:

Method 1: obtaining equal prediction steps in the horizontal direction and the vertical direction according to the size of the current image block;

For example, if the current image block size is 64×64, the horizontal and vertical directions are 4:1 to obtain the predicted pixel value, then the predicted pixel values are obtained. The number is 16 x 16.

Manner 2: The non-equal step size in the horizontal direction and the vertical direction is obtained according to the size of the current image block to obtain a predicted pixel value.

The step size in the horizontal direction is 2, and the step size in the vertical direction is 4 to obtain the predicted pixel value. For example, for the current image block size of 16 x 16 , the horizontal direction is 2:1 to obtain the horizontal direction. The predicted pixel value on the upper side is 4:1 in the vertical direction to obtain the predicted pixel value in the vertical direction, and the number of predicted pixel values obtained is 8×4.

Method 3: Determine the step size according to the size of the current image block, and obtain the predicted pixel value according to the step size. Taking 32 X 32 and 64 X 64 image blocks as an example, let 32 χ 32 image blocks have a horizontal and vertical step size of 2, and 32 x 32 image blocks have a horizontal and vertical step size of 4, for these two sizes. The number of predicted pixel values obtained by the image block is 16 X 16 .

Correspondingly, the decoding end can perform decoding processing by using a processing procedure corresponding to the encoding end. Optionally, the foregoing process of obtaining the predicted pixel value in a preset step size may be preset by the encoding end and the decoding end, or may not be preset.

When it is not preset by the encoding end and the decoding end, for example, the encoding end compares the processing results of the values of the respective step sizes to determine an optimal step size for processing, and at the time of encoding, The value of the preset step size is encoded, so that the decoding end performs corresponding processing according to the value of the step size of the identifier.

S304: Calculate a first mean value of the prediction data according to the predicted pixel value, and calculate a second mean value of the data of the current image block according to the pixel value of the data of the current image block. S305: Perform a difference between the second mean value of the data of the current image block and the first mean value of the prediction data of the current image block to obtain a residual of the current image block.

S306: Encode the residual to obtain a coding result of the current image block.

In this step, optionally, the encoding process of the residual includes at least the following implementation manners: The first implementation manner: quantizing the residual, quantizing the residual to a smaller value, and using less The bit is encoded, and accordingly, at the decoding end, the inverse quantization process is performed using the same quantization scale, and the result of the inverse quantization process is used as the residual used.

The second implementation manner is: directly encoding the residual, and correspondingly, the decoding end decodes the residual, and uses the decoding processing result as the residual used to avoid the compression loss caused by the quantization.

In a third implementation manner, the residual is mapped to another residual, and another residual is encoded, indicating that the bit of another residual is smaller than the bit representing the residual, thereby further improving coding in a specific implementation. It can be mapped according to the mapping table. For example, if the pixel values in the depth image are 8, 64, 164, 240, 245, then the pixel values in the corresponding mapping table are 1, 2, 3, 4, 5, and in this way, the coding bits can be reduced to further Improve coding efficiency.

S307: Other processing.

For example, the residual matrix obtained by performing the point-by-point difference between the data of the current image block and the pixel value of the reference data of the current image block may be encoded to obtain a coding result, or the method of Wedgelet may be used. A residual matrix obtained by subtracting a constant value from each pixel in the current image block is subjected to an encoding process to obtain a coding result, which is not described in detail in this embodiment.

In this embodiment, by obtaining the data of the current image block of the depth image, obtaining the predicted pixel value in the data of the current image block according to the preset step size, obtaining the prediction data corresponding to the current image block of the depth image, according to the preset step size. Obtaining a predicted pixel value in the prediction data, calculating a first mean value of the prediction data according to the predicted pixel value, calculating a second mean value of the data of the current image block according to the pixel value of the data of the current image block, and secondizing the data of the current image block The mean value is different from the first mean value of the prediction data of the current image block, and the residual of the current image block is obtained, and the residual is encoded to obtain the coding result of the current image block, which can reduce the reference pixel of the prediction data of the current image block. The number of points and the number of reference pixels of the data of the current image block, so that when calculating the first mean value of the prediction data of the current image block and the second mean value of the data of the current image block, the calculation amount is significantly reduced. Thereby, the coding efficiency can be improved. Correspondingly, when the decoding end decodes the current image block, The calculation amount of the decoding process by the decoding method corresponding to the encoding method of the encoding side is also significantly reduced. Therefore, the coding and decoding efficiency can be effectively improved by using the technical solution of the embodiment.

4 is a schematic flowchart of a first embodiment of a method for decoding a depth image according to the present invention. As shown in FIG. 4, the execution body of the embodiment is a decoding device, and the method of this embodiment includes:

S401: Obtain prediction data corresponding to the current image block of the depth image, obtain a predicted pixel value in the prediction data according to the preset step size, and calculate a first mean value of the prediction data according to the predicted pixel value, where the preset step size is not positive Integer

In this embodiment, the preset step size may be used to identify an interval between predicted data position coordinates. Specifically, the decoding end can still use the process corresponding to the above encoding end to implement the step of obtaining the predicted pixel value in the prediction data according to the preset step size.

For example, if the size of the prediction data of the current image block is 8 x 8, that is, there are 64 pixel points. With a preset step size of 2, the predicted data of the current image block can be processed 2:1, that is, one pixel is spaced once in the horizontal direction and the vertical direction, and one pixel point is obtained. The predicted pixel value can be obtained.

S402. Obtain a residual of the current image block.

Specifically, the encoding end may obtain prediction data corresponding to the current image block of the depth image, obtain a predicted pixel value in the prediction data according to the preset step size, calculate a first mean value of the prediction data according to the predicted pixel value, and according to the predicted data. The first mean value and the pixel value of the pixel of the current image block are obtained as residuals, and the residual is encoded and sent to the decoding end, and the decoding end can acquire the residual of the current image block from the code stream.

S403. Obtain a pixel value of a pixel of the current image block according to the first mean value of the prediction data and the residual of the current image block.

In this embodiment, by obtaining the prediction data corresponding to the current image block of the depth image, obtaining the predicted pixel value in the prediction data according to the preset step size, and calculating the first mean value of the prediction data according to the predicted pixel value, the preset step size is A positive integer not being 1, obtaining a residual of the current image block, obtaining a pixel value of a pixel of the current image block according to the first mean value of the prediction data and the residual of the current image block, and reducing the reference pixel of the prediction data of the current image block The number of points, thus, when calculating the first mean value of the prediction data of the current image block, the calculation amount is significantly reduced, so that the decoding efficiency can be improved.

In the foregoing embodiment, when S403 is specifically implemented, for example, the following two methods may be implemented: Method 1: Adding a first mean value of prediction data of a current image block to a residual of a current image block, Obtaining a pixel value of a pixel of the current image block;

Corresponding to the method, the processing of the encoding end is to make the difference between the pixel point of the current image block and the first average value of the prediction data of the current image block to obtain the residual of the current image block, and then the decoding end can perform the current image block. The first mean of the prediction data is added to the residual of the current image block to obtain the pixel value of the pixel of the current image block.

Manner 2: adding the first mean value of the prediction data of the current image block to the residual, obtaining a pixel mean value of the current image block, and using the pixel mean value of the current image block as the pixel value of the pixel point of the current image block.

Corresponding to the method, the processing of the encoding end is to obtain the data of the current image block of the depth image, obtain the predicted pixel value in the data of the current image block according to the preset step size, and calculate the current image according to the pixel value of the data of the current image block. The second mean value of the data of the block is obtained by comparing the second mean value of the data of the current image block with the first mean value of the predicted data of the current image block to obtain a residual of the current image block, and the processing process corresponding to the decoding end is The first mean value of the prediction data of the image block is added to the residual, and the pixel mean of the current image block is obtained, and the pixel mean of the current image block is taken as the pixel value of the pixel point of the current image block.

The decoding method of the present invention will be described in detail below using a specific embodiment.

FIG. 5 is a schematic flowchart of a second embodiment of a method for decoding a depth image according to the present invention. The execution body of the embodiment is a decoding device, and the method in this embodiment includes:

S501: Determine, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set; if yes, execute S502~S506, if not, execute S507.

S502: Determine to perform a sample processing on the reference image block corresponding to the current image block.

The intra prediction mode set in this embodiment includes at least a DC mode, a Planar mode, a displayed identification mode, and an intra prediction mode based mode, and the reference image block corresponding to the current image block is performed according to a preset lowering rule. Before the processing is performed, it is determined whether the intra prediction mode belongs to the intra prediction mode set, and accordingly, the encoding end has the same intra prediction mode set as the decoding end, and corresponds to the current image block and the current image block. Before the reference image block is subjected to the sample processing, it is also necessary to first determine whether the intra prediction mode belongs to the combination of the intra prediction modes.

The preset intra prediction mode set is a set of intra prediction modes that the encoding end has been tested to show that the encoding method is averaged to obtain better encoding processing performance.

If it belongs, it is determined that the reference image block corresponding to the current image block is subjected to the sample processing. It should be noted that steps S501 and S502 are optional steps, and may not be performed, and S503~S506 are directly executed.

S503: Obtain prediction data corresponding to the current image block of the depth image, and obtain a predicted pixel value in the prediction data according to the preset step size.

Obtaining predicted pixel values in the prediction data according to the preset step size can be achieved in the following three ways:

Optionally, the foregoing process of obtaining the predicted pixel value in a preset step size may be preset by the encoding end and the decoding end, or may not be preset.

When the encoding end and the decoding end are not preset, for example, the encoding end compares the processing results of the values of the respective step sizes, and determines an optimal step size for processing. Therefore, when encoding, the encoding end needs to be The value of the preset step size is encoded, and the decoding end can parse the value of the step size of the identifier from the code stream.

S504: Calculate a first mean value of the prediction data according to the predicted pixel value of the prediction data.

Since the number of predicted pixel values of the prediction data is smaller than the number of pixels of the prediction data, the amount of calculation of step 504 is reduced, and the decoding efficiency can be improved. S505: Obtain a residual of the current image block.

Specifically, the following implementation manners may be used in this step, and each implementation manner corresponds to the implementation manner of the encoding end:

The first implementation manner: 反 The inverse quantization process is performed on the residual by the same quantization method as the coding end, and the result of the inverse quantization process is used as the residual used.

The second implementation manner: When the encoding end directly encodes the residual, the decoding end decodes the residual, and uses the decoding processing result as the residual used to avoid the compression loss caused by the quantization.

The third implementation manner is: when the coding end maps the residual to the residual mapping value, the residual mapping value is encoded, and the bit indicating the residual mapping value is smaller than the bit representing the residual, thereby further improving the coding. effectiveness. Then, the decoding end can also reflect the coding result in a corresponding manner.

S506: Add a first mean value of the prediction data of the current image block to a residual of the current image block to obtain a pixel value of a pixel point of the current image block.

This step is an implementation process of the first method of the foregoing S403. It can be understood that if the encoding end uses the encoding process corresponding to the second mode, the step can also be implemented by the second method of the above S403.

S507: Other processing.

For other processing, for example, corresponding to the encoding end, the decoding end decodes the encoding result to obtain a residual matrix, and sums the pixel points of the residual matrix and the reference image block point by point to obtain a sum value matrix, and restores the current matrix according to the sum value matrix. The image block, or, when the encoding end uses the Wedgelet method, encodes a residual matrix obtained by subtracting each pixel point in the current image block from a constant value, and obtains a coding result, and the decoding end performs the coding result. The decoding process is performed to obtain a residual matrix, and each pixel value of the residual matrix is summed with a constant value to obtain a sum matrix. The current image block is restored according to the sum value matrix, which is not described in detail in this embodiment.

In this embodiment, by obtaining the prediction data corresponding to the current image block of the depth image, obtaining the predicted pixel value in the prediction data according to the preset step size, calculating the first mean value of the prediction data according to the predicted pixel value of the prediction data, and obtaining the current image. The residual of the block, adding the first mean value of the prediction data of the current image block to the residual of the current image block, obtaining the pixel value of the pixel point of the current image block, and reducing the reference pixel point of the prediction data of the current image block. Number, thus, the current image is obtained in the calculation When the first mean value of the block prediction data is calculated, the amount of calculation is significantly reduced, so that the decoding efficiency can be improved. It should be noted that the technical solution of the embodiment of the present invention is also applicable to other intra prediction methods that need to encode and decode the mean. For example, the coding end encodes the residual by performing a difference between the mean value of the current image block and a constant value, and obtains the coding result, and the decoding end performs decoding processing on the coding result to obtain a residual, and the residual and the constant are constant. The sum of the values is used as the pixel value of each pixel of the current image block, and the current image block is restored. When the current image block is averaged at the encoding end, the current image block may be predicted according to the preset step size. The pixel value is averaged for the predicted pixel value of the current image block, and the residual value is compared with a constant value to obtain a residual, and the residual is encoded to obtain a coding result. Correspondingly, the decoding end may also use a similar method for decoding. Processing, thereby improving the codec efficiency.

FIG. 6 is a schematic structural diagram of Embodiment 1 of a depth image encoding apparatus according to the present invention. As shown in FIG. 6, the apparatus in this embodiment includes a first processing module 61, a second processing module 62, and an encoding module 63, where the first processing The module 61 is configured to obtain prediction data corresponding to the current image block of the depth image, obtain a predicted pixel value in the prediction data according to the preset step size, and calculate a first mean value of the prediction data according to the predicted pixel value, where the preset step size is not 1 The second processing module 62 is configured to obtain a residual according to the first mean value of the prediction data and the pixel value of the pixel point of the current image block; the encoding module 63 is configured to encode the residual of the current image block.

The coding device of this embodiment can be used to implement the technical solution of the method embodiment shown in FIG. 1. The principle and technical effects are similar, and details are not described herein again.

In the above embodiment, the second processing module 62 is specifically configured to compare the pixel of the current image block with the first average of the prediction data of the current image block to obtain a residual of the current image block; or obtain a current image of the depth image. Block data, obtaining predicted pixel values in data of the current image block according to a preset step size, calculating a second mean value of data of the current image block according to predicted pixel values of data of the current image block, and data of the current image block The second mean is different from the first mean of the prediction data of the current image block to obtain a residual of the current image block.

In the above embodiment, the first processing module 61 is specifically configured to obtain the predicted pixel values according to the equal size of the horizontal direction and the vertical direction according to the size of the current image block; or, perform horizontal and vertical directions according to the size of the current image block. The non-equal step size obtains the predicted pixel value.

In the above embodiment, the encoding module 63 is further configured to encode the value of the preset step size. In the above embodiment, the preset step size is used to identify the interval between the predicted data position coordinates. In the foregoing embodiment, the encoding module 63 is specifically configured to map the residual into a residual mapping value, and the bit representing the residual mapping value is smaller than the bit representing the residual, and the residual mapping value is encoded.

In the foregoing embodiment, the first processing module 61 is further configured to determine, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set, and if yes, determine according to the preset step size. The predicted pixel value is obtained in the prediction data.

In the above embodiment, the intra prediction mode set includes at least one of the following intra prediction modes: DC DC mode; Planar Planar mode; explicitly identified Wedgelet mode; Wedgelet mode based on intra prediction mode.

The coding device of this embodiment can be used to implement the technical solution of the method embodiment shown in FIG. 3, and the principle and the technical effect are similar, and details are not described herein again.

FIG. 7 is a schematic structural diagram of Embodiment 1 of a decoding apparatus for a depth image according to the present invention. As shown in FIG. 7, the apparatus of this embodiment includes: a processing module 71 and a decoding module 72, where the processing module 71 is configured to obtain a current image of a depth image. The prediction data corresponding to the block obtains a predicted pixel value in the prediction data according to a preset step size, and calculates a first mean value of the prediction data according to the predicted pixel value, where the preset step size is a positive integer that is not 1; the decoding module 72 is configured to: Obtaining a residual of the current image block, and obtaining a pixel value of a pixel point of the current image block according to the first mean value of the prediction data and the residual of the current image block.

The coding device of this embodiment can be used to implement the technical solution of the method embodiment shown in FIG. 4, and the principle and the technical effect are similar, and details are not described herein again.

In the above embodiment, the decoding module 72 is specifically configured to add the first mean value of the prediction data of the current image block to the residual of the current image block to obtain the pixel value of the pixel point of the current image block; or, the current image block The first mean value of the predicted data is added to the residual, and the pixel mean of the current image block is obtained, and the pixel mean of the current image block is taken as the pixel value of the pixel of the current image block.

In the above embodiment, the processing module 71 is specifically configured to obtain the predicted pixel value according to the equal size of the horizontal direction and the vertical direction according to the size of the current image block; or, perform the horizontal and vertical unequal according to the size of the current image block. The step size gets the predicted pixel value.

In the above embodiment, the decoding module 72 is further configured to parse the code stream to obtain a value of the identification step size.

In the above embodiment, the preset step size is used to identify the interval between the predicted data position coordinates. In the foregoing embodiment, the decoding module 72 is specifically configured to perform decoding processing on the encoding result of the current image block, obtain a residual mapping value of the current image block, map the residual mapping value into a residual, and use the residual as the current image block. The residual of the residual map value is smaller than the bit representing the residual. In the above embodiment, the processing module 71 is further configured to determine, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set, and if yes, determine that the prediction is based on the preset step size. The predicted pixel value is obtained in the data.

The coding device of this embodiment can be used to implement the technical solution of the method embodiment shown in FIG. 5. The principle and technical effects are similar, and details are not described herein again.

FIG. 8 is a schematic flowchart of Embodiment 3 of a method for encoding a depth image according to the present invention. As shown in FIG. 7, the method in this embodiment includes:

S801: Obtain prediction data corresponding to the current image block of the depth image.

Specifically, corresponding prediction data may be obtained according to a prediction mode of the current image block.

S802: Perform an arithmetic operation on every N pixel points in the pixel in the prediction data to obtain an arithmetic value of the corresponding pixel value.

As a feasible implementation, an average operation is performed for every N pixels in the pixel in the prediction data to obtain an average pixel value, and the average pixel value is used as an arithmetic value of the corresponding pixel.

The arithmetic value of the corresponding pixel value can also be obtained by performing other pixel operations on every N pixel points in the pixel in the prediction data, which is not limited in the present invention.

Accordingly, the decoding terminal obtains the arithmetic value of the corresponding pixel value by the same arithmetic operation as the encoding end. S803: Calculate a first mean value of the prediction data according to an arithmetic value of the corresponding pixel value.

S804: Obtain a residual according to the first mean value of the prediction data and the pixel value of the pixel of the image block; optionally, as a feasible implementation manner, the pixel point of the image block is the first of the prediction data of the current image block. The mean is poor and the residual of the current image block is obtained.

As another feasible implementation manner, the data of the current image block of the depth image is obtained, the predicted pixel value is obtained in the data of the current image block according to the preset step size, and the second mean value of the predicted data is calculated according to the predicted pixel value, and the image is obtained. The second mean of the block is compared to the first mean of the reference image block of the current image block to obtain the residual of the current image block.

S805: Encode the residual of the current image block.

In this embodiment, by obtaining the prediction data corresponding to the current image block of the depth image, performing arithmetic operations on each N pixel points in the pixel in the prediction data, and obtaining an arithmetic value of the corresponding pixel value, according to An arithmetic value of the corresponding pixel value calculates a first mean value of the prediction data; a residual is obtained according to a first mean value of the prediction data and a pixel value of a pixel point of the image block; and a residual of the current image block is encoded. Since arithmetic operations are performed on every N pixel points in the pixel in the prediction data, the arithmetic value of the corresponding pixel value is obtained, and a better pixel value is obtained, thereby improving the encoding and decoding efficiency.

FIG. 9 is a schematic flowchart of Embodiment 3 of a method for decoding a depth image according to the present invention. As shown in FIG. 8, the method in this embodiment includes:

S901: Obtain prediction data corresponding to the current image block of the depth image.

S902: Perform an arithmetic operation on every N pixel points in the pixel in the prediction data to obtain an arithmetic value of the corresponding pixel value.

The arithmetic operation used by the decoding end obtains the arithmetic value of the corresponding pixel value as the encoding end.

S903: Calculate a first mean value of the prediction data according to an arithmetic value of the corresponding pixel value.

S904: Obtain a residual of the current image block;

S905: Obtain a pixel value of a pixel of the current image block according to the mean value of the prediction data and the residual of the current image block.

Optionally, as a feasible implementation manner, the mean value of the prediction data of the current image block is added to the residual of the current image block to obtain a pixel value of a pixel point of the current image block.

As another feasible implementation manner, the mean value of the prediction data of the current image block is added to the residual, and the pixel mean of the current image block is obtained, and the pixel average of the current image block is taken as the pixel value of the pixel of the current image block.

In this embodiment, by obtaining the prediction data corresponding to the current image block of the depth image, performing arithmetic operations on each N pixel points in the pixel in the prediction data, obtaining an arithmetic value of the corresponding pixel value, and calculating according to the arithmetic value of the corresponding pixel value. Predicting a first mean value of the data; obtaining a residual of the current image block; obtaining a pixel value of a pixel of the current image block according to the mean of the predicted data and the residual of the current image block. The residual of the current image block is encoded. Since arithmetic operations are performed on every N pixel points in the pixels in the prediction data, the arithmetic value of the corresponding pixel value is obtained, and the codec efficiency is improved. FIG. 10 is a schematic structural diagram of a second embodiment of an apparatus for encoding a depth image according to a second embodiment of the present invention. The apparatus of the present embodiment includes: a first processing module 1001, a second processing module 1002, and an encoding module 1003. The first processing module 1001 is configured to obtain prediction data corresponding to the current image block of the depth image, perform an arithmetic operation on each N pixel points in the pixel in the prediction data, obtain an arithmetic value of the corresponding pixel value, and perform arithmetic according to the corresponding pixel value. The value calculates a first mean value of the prediction data; the second processing module 1002 is configured to obtain a residual according to the first mean value of the prediction data and the pixel value of the pixel point of the current image block; the encoding module 1003 is configured to perform the residual of the current image block. coding.

In the above embodiment, the first processing module 1001 is specifically configured to perform an averaging operation on every N pixels in the pixel in the prediction data to obtain an average pixel value, and use the average pixel value as an arithmetic value of the corresponding pixel.

In the above embodiment, the second processing module 1002 is specifically configured to compare the pixel of the image block with the first average of the prediction data of the current image block to obtain a residual of the current image block; or, obtain a current image block of the depth image. Data, obtaining a predicted pixel value in the data of the current image block according to a preset step size, calculating a second mean value of the predicted data according to the predicted pixel value, and comparing the second mean value of the image block with the reference image block of the current image block A mean value is made to get the residual of the current image block.

The coding device of this embodiment can be used to implement the technical solution of the method embodiment shown in FIG. 8. The principle and technical effects are similar, and details are not described herein again.

11 is a schematic structural diagram of Embodiment 2 of a decoding apparatus for a depth image according to the present invention. The apparatus of this embodiment includes a processing module 1101, and a decoding module 1102, where the processing module 1101 is configured to obtain prediction data corresponding to a current image block of a depth image. An arithmetic operation is performed every N pixels in the pixel in the prediction data, an arithmetic value of the corresponding pixel value is obtained, and a first mean value of the prediction data is calculated according to an arithmetic value of the corresponding pixel value; and the decoding module 1102 is configured to obtain the residual of the current image block. Poor, the pixel value of the pixel of the current image block is obtained according to the mean of the prediction data and the residual of the current image block.

In the above embodiment, the processing module 1101 is specifically configured to perform an averaging operation on every N pixel points in the current image block and the prediction data to obtain an average pixel value, and use the average pixel value as an arithmetic value of the corresponding pixel value.

In the above embodiment, the decoding module 1102 is specifically configured to add the average value of the prediction data of the current image block to the residual of the current image block to obtain the pixel value of the pixel of the current image block; or, predict the current image block. The mean value of the data is added to the residual, and the pixel mean of the current image block is obtained, and the pixel mean of the current image block is taken as the pixel value of the pixel of the current image block. The coding device of this embodiment can be used to implement the technical solution of the method embodiment shown in FIG. 9. The implementation principle and technical effects are similar, and details are not described herein again. In each of the above embodiments, each of the average values may be represented as a divisor of a sum of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data, and the sum of the pixel values of the pixel points in the prediction data is sum. , the division value of the pixel value in the prediction data is numOfPixel, then the mean value is expressed as sum I numOfPixelo

In addition, the method of obtaining the mean can be obtained by the following steps:

Obtaining a sum value of pixel values of pixel points in the prediction data and a number of pixel points in the prediction data;

Determining whether the value of the pixel in the predicted data is 0;

If the value of the pixel in the prediction data is not 0, the ninth value is divided with the value of the pixel point in the prediction data to obtain the mapping value;

And multiplying the mapping value by a sum of pixel values of pixel points in the prediction data to obtain a first value; performing a shift operation on the first value to obtain a second value;

If the value of the pixel in the prediction data is 0, a preset constant is taken as the second value; the second value is taken as the mean value. According to the above method, the ninth value is a fixed value or the fixed value plus an offset R, the fixed value is 2, the offset is half of the value of the pixel point in the prediction data, or the offset is The value of the pixel point in the prediction data is shifted to the right by one. Where N can take a value of 14 or other positive integer. According to the above method,

The dividing the ninth value and the value of the pixel in the prediction data, and obtaining the mapping value includes:

Dividing 2 or ι ^Ν by R by the value of the pixel in the prediction data to obtain the mapping value; performing a shift operation on the first value to obtain the second value includes:

The first value is shifted to the right by N bits to obtain a second value. According to the above method, the ninth value is divided with the value of the pixel in the prediction data Operation, get the mapping value including:

And shifting the value of the pixel in the prediction data to the right by M bits, and obtaining the value of the pixel in the shifted prediction data;

Dividing 1 ^N plus R by the value of the pixel in the shifted prediction data to obtain the mapping value;

The shifting operation is performed on the first value, and obtaining the second value includes:

The first value is shifted right by N plus M bits to obtain a second value. According to the above method, the ninth value is divided with the value of the pixel in the prediction data, and before the mapping value is obtained, the method further includes:

Determining whether the value of the pixel in the predicted data is greater than a preset value;

If it is greater than, the ninth value is divided by the value of the pixel in the prediction data, and the obtained mapping value includes:

Dividing ^N or R by the value of the pixel in the shifted prediction data to obtain the mapping value;

The first value is shifted right by N plus M bits to obtain a second value. Where M can have a value of 3 or other positive integer.

If it is smaller, the ninth value is divided with the value of the pixel in the prediction data, and the obtained mapping value includes:

The first value is shifted to the right by N bits to obtain a second value. The method of a specific embodiment includes:

SS101: Acquire prediction data of the current image block of the depth image.

The prediction data of the current image block of the depth image is related to the coding mode used. For example, the prediction data is the corresponding prediction data obtained by using the intra prediction mode, and the size and current picture of the prediction data. The size of the block is the same.

The current image block may be the current intra prediction block. In addition, since the prediction block and the transform block are bound in the intra prediction, the intra prediction is also usually performed in units of transform blocks; at this time, the image block is a transform. Piece.

SS102: Obtain a sum value of pixel values of pixel points in the prediction data and a value of the pixel point in the prediction data.

SS103: Determine whether the value of the pixel in the prediction data is 0. If not, execute SS104-SS106, if SS107 is executed.

SS104: dividing a ninth value and a value of a pixel in the prediction data to obtain a mapping value;

Wherein, the ninth value is a fixed value or a fixed value plus an offset R, R is half of the number of pixels in the prediction data, and the mapping value is twice the reciprocal of the value of the pixel in the prediction data, and the above is greater than the prediction The value of a pixel in the data, N is an integer.

SS105: Multiply the map value by the sum of the pixel values of the pixel points in the prediction data to obtain the first value.

SS106: performing a shift operation on the first value to obtain a second value;

Since the mapping value is 2^ times the reciprocal of the value of the pixel in the prediction data, and multiplied by the mapping value of the value of the pixel in the prediction data, the obtained first numerical value directly compares the pixel in the prediction data. The result of dividing the sum of the pixel values by the value of the pixel is increased by a factor of 2. Therefore, the second value can be obtained by the removal operation of the SS 106, so that the second value is directly compared with the pixel of the pixel in the predicted data. The sum of the values is equal to the result of dividing the values of the pixels.

SS107: A preset constant is used as the second value.

That is, when there is no pixel in the prediction data, a preset constant is used as the second value, and the preset constant is generally 0 or 128 left shift bit-8, where bit represents the bit width of the pixel in the image.

SS108: The second value is taken as the mean. In the above SS108, if the encoding method involved is also obtained by averaging the current image block, it can be obtained in a similar manner to the second image value obtained by the prediction image block, and the present invention will not be described in detail.

In this embodiment, by dividing the ninth value and the value of the pixel in the prediction data, the mapping value is obtained, and the mapping value is multiplied by the sum of the pixel values of the pixel points in the prediction data to obtain the first value. And then shifting the first value to obtain the second value, and the second value The pixel value of the pixel of the front image block obtains the residual of the current image block, and encodes the residual of the current image block, and the complexity of the implementation in the encoding process is only related to the value range of the pixel value in the prediction data. Therefore, the complexity of the implementation is reduced to some extent.

2 is a schematic flowchart of a second embodiment of a method for encoding a depth image according to the present invention. The execution body of the embodiment is an encoding device. This embodiment is a specific implementation manner in the case where the value of the pixel in the prediction data is not zero. The method of this embodiment includes:

SS201: Acquire prediction data of the current image block of the depth image.

This step is similar to SS101 and will not be described here.

SS202: Obtain a sum of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data.

SS203: Divide 2^ by the value of the pixel in the prediction data to obtain the mapped value.

Wherein, the mapping value is twice the reciprocal of the numerical value of the pixel in the prediction data, and the above 2 is larger than the numerical value of the pixel in the prediction data, and N is an integer.

The reason why this step is to divide 2 by the number of pixels in the predicted data is to avoid floating point numbers.

SS204: Multiply the mapped value by the sum of the pixel values of the pixel points in the predicted data to obtain the first value.

SS205: Move the first value to the right by N bits to obtain the second value.

Since the mapping value is 2^ times the reciprocal of the value of the pixel in the prediction data, and multiplied by the mapping value of the value of the pixel in the prediction data, the obtained first numerical value directly compares the pixel in the prediction data. The sum of the pixel values is divided by the value of the pixel by 2^ times. Moving a value to the right by N bits in binary is equivalent to reducing the value by 2^ times, shifting a value to the left by N bits. , equivalent to increasing the value by a factor of two. Therefore, the second value can be obtained by shifting the first value to the N position by the SS205, so that the second value is directly compared with the pixel value of the pixel point in the prediction data. The result of dividing the values of the points is equal.

SS206: The second value is taken as the mean.

The residual of the current image block is encoded, and the encoded code stream is sent to the decoding end.

It should be noted that, in the above embodiment, the case where the value of the pixel point in the prediction data is 0 is the same as the processing method in which the value of the pixel point in the prediction data in the embodiment shown in FIG. 1 is 0. No longer. In this embodiment, the map value is obtained by dividing the value of the pixel point in the prediction data. Multiplying the mapped value by the sum of the pixel values of the pixel points in the predicted data to obtain a first value, and then shifting the first value to the right by N bits to obtain a second value, and the second value is compared with the pixel of the current image block. The pixel value obtains the residual of the current image block, and encodes the residual of the current image block. In the process of acquiring only the mapped value in the encoding process, the dividing operation is involved, and dividing 2 by the value of the pixel in the predicted data, The mapping value, 2^ is a fixed value. Therefore, the complexity of the implementation is only related to the value range of the pixel values in the prediction data, which reduces the complexity of the implementation to a certain extent, reduces the chip area, and improves the processing. speed.

On the basis of the foregoing embodiments, after acquiring the values of the pixel points in the prediction data, shifting the values of the pixel points in the prediction data to the right by M bits to reduce the number of pixels in the prediction data. The value range of the value is further shifted by the first numerical value so that the second value is equal to the result of directly dividing the sum of the pixel values of the pixel points in the prediction data by the value of the pixel point, thereby further reducing the implementation. The method of the embodiment specifically includes:

SS301: Acquire prediction data of the current image block of the depth image.

This step is similar to SS101 and will not be described here.

SS302: Obtain a sum of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data.

SS303: The value of the pixel in the prediction data is shifted to the right by M bits, and the value of the pixel in the shifted prediction data is obtained.

The value of the pixel in the prediction data is shifted to the right by M bits, which is equivalent to reducing the value of the pixel in the prediction data to the original 1/2 ^M , where M is a positive integer.

SS304: Divide 2^ by the value of the pixel in the shifted prediction data to obtain the mapped value. The reason why this step is to divide 2 by the value of the pixel in the prediction data is to avoid floating point number.

SS305: Multiply the mapped value by the sum of the pixel values of the pixels in the predicted data to obtain the first value.

SS306: Move the first value to the right by N plus M bits to obtain the second value.

Since the value of the pixel in the prediction data is first shifted to the right by M bits, and then 2^ is divided by the value of the pixel in the shifted prediction data to obtain a mapping value, which is compared with the pixel in the prediction data. After multiplying the mapped values of the values, the first value obtained is directly compared to the pixels of the pixels in the predicted data. The result of dividing the sum of the values by the value of the pixel is increased by 2^ 咅. Therefore, the second value can be obtained by shifting the first value to the right by N+M bits of SS306, so that the second value is directly predicted. The sum of the pixel values of the pixels in the data is equal to the result of dividing the values of the pixels.

SS308: Use the second value as the mean. In this embodiment, by first shifting the values of the pixel points in the prediction data to the right by M bits, the values of the pixel points in the shifted prediction data are obtained, and then 1 ^{N is} divided by the pixel points in the shifted prediction data. a value, obtain a mapped value, multiply the mapped value by the sum of the pixel values of the pixel points in the predicted data to obtain a first value, and shift the first value to the right by N plus M bits to obtain a second value, which will be the second The value and the pixel value of the pixel of the current image block obtain the residual of the current image block, and encode the residual of the current image block. In the encoding process, the value of the pixel in the prediction data is shifted to the right by M bit, by 2 Dividing by the value of the pixel in the shifted prediction data, the mapping value is obtained, and only 2 is divided by the value of the pixel in the shifted prediction data, and the division operation is involved in the process of obtaining the mapping value, and 2 is A fixed value, therefore, the complexity of the implementation is only related to the range of values of the pixels in the shifted prediction data, further reducing the complexity of the implementation, reducing the chip area, and increasing the processing speed.

In order to further reduce the complexity of the implementation, the method in this embodiment specifically includes:

SS401: Acquire prediction data of the current image block of the depth image.

SS402: Obtain a sum of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data.

SS403: Determine whether the value of the pixel in the predicted data is greater than the preset value. If yes, execute SS404-SS407, if not, execute SS408~SS410.

SS404: The value of the pixel in the prediction data is shifted to the right by M bits, and the value of the pixel in the shifted prediction data is obtained.

SS405: Divide 2^ by the value of the pixel in the shifted prediction data to obtain a mapped value.

SS406: multiplying the mapped value by a sum of pixel values of pixel points in the prediction data to obtain a first value;

SS407: Move the first value to the right by N plus M bits to obtain the second value.

The above SS404~SS407 are similar to the corresponding steps in FIG. 3, and details are not described herein again.

SS408: Divide 2^ by the value of the pixel in the prediction data to obtain a mapped value;

SS409: Multiply the mapped value by the sum of the pixel values of the pixels in the predicted data to obtain the first number Value

SS410: Move the first value to the right by N bits to obtain the second value.

The foregoing SS408~SS410 are similar to the corresponding steps in FIG. 2, and are not described herein again.

SS411 : The second value is taken as the mean value.

In the above steps, the complexity of the implementation may be reduced to some extent due to the addition of the judgment condition. For example, when the preset value is 2 1 ^ΰ , when the pixel value of the pixel of the prediction data is 2 ⁸ , divide by 2 and divide by 2 2 ⁸ , obtain a mapping value, multiply the mapping value by the sum of the pixel values of the pixel points in the prediction data, obtain the first value, and shift the first value to the right to obtain the second value. When the pixel value of the pixel of the prediction data is 2 ¹² , the 2 ^{12 is} shifted to the right by 4 bits, the shift is 2 ⁸ , and the division is 2 ⁸ to obtain a mapping value, and the mapping value and the prediction data are The sum of the pixel values of the pixels is multiplied to obtain a first value, and the first value is shifted right by N+M bits to obtain a second value. It can be seen that the division operation for the values of the pixel points in the two different prediction data in the above process is the same as the division operation of the values of one pixel point. Therefore, the complexity of the implementation can be further reduced by the method of the present embodiment.

It should be noted that the above steps SS408~SS410 are processing methods when the value of the pixel in the prediction data is less than the preset value and is not 0. If the value of the pixel in the prediction data is 0, It is the same as the processing method in which the value of the pixel in the prediction data in the foregoing embodiment is 0, and details are not described herein again.

In this embodiment, after obtaining the value of the pixel point in the prediction data, adding a judgment condition, determining whether the value of the pixel point in the prediction data is greater than a preset value, when the value of the pixel point in the prediction data is greater than the pre-predetermined value When the value is set, the operation corresponding to the numerical shift of the pixel points is used, and the complexity of the implementation is only related to the value range of the pixel values in the shifted prediction data, which can further reduce the complexity of the implementation. When the value of the pixel in the prediction data is smaller than the preset value, the operation of not shifting the value of the pixel point is used, and the precision of the division can be ensured; the complexity of the implementation is further reduced by increasing the judgment condition.

In addition, in the above embodiments, there is a certain error due to the division operation of the values of the pixels in the prediction data. Considering that the data in the above embodiment has a certain range of values, for example, the mean range is 0 ~ 255. In order to ensure that the result of the above embodiment is still within the preset range, the result obtained by the above embodiment can be subjected to a clip operation to ensure that the data is within a preset range. In addition, the mean can be obtained as follows:

Obtaining a mapping value of a value of a pixel in the predicted data;

Multiplying the map value by a sum of pixel values of pixel points in the prediction data; obtaining a fifth value; performing a shift operation on the fifth value to obtain an eighth value;

This sixth value is taken as the mean value. According to the above method,

The mapping values for obtaining the values of the pixel points in the prediction data include:

Obtaining a mapping value corresponding to the value in the lookup table according to the value of the pixel in the prediction data;

The shifting operation is performed on the fifth value, and obtaining the sixth value includes:

The fifth value is shifted right by N bits to obtain a sixth value. Where N can be 14 or other positive integer. According to the above method,

The value of the pixel in the prediction data is shifted to the right by M bits, and the value of the pixel in the shifted prediction data is obtained; where M can be 3 or other positive integer.

Obtaining a mapping value corresponding to the value in the lookup table according to the value of the pixel point in the shifted prediction data;

The fifth value is shifted right by N plus M bits to obtain a sixth value. According to the above method, after obtaining the value of the pixel in the prediction data, the method further includes: determining whether the value of the pixel in the prediction data is greater than a preset value;

If it is greater than, the mapping value of obtaining the value of the pixel in the prediction data includes:

Obtaining the pair of values in the lookup table according to the values of the pixel points in the shifted prediction data. The mapped value

The fifth value is shifted right by N plus M bits to obtain a sixth value;

If it is smaller, the mapping value of obtaining the value of the pixel in the prediction data includes:

Performing a shift operation on the fifth value to obtain a sixth value;

The fifth value is shifted right by N bits to obtain a sixth value. The specific method of this embodiment includes:

S1001: Obtain prediction data of the current image block of the depth image.

S1002: Obtain a sum of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data.

S1003: Obtain a mapping value corresponding to a value in the lookup table according to the value of the pixel in the prediction data.

The lookup table contains two sets of data, which are mapping values corresponding to the values of the pixels in the predicted data and the values of the pixels in the predicted data. When the value of the pixel in the predicted data is not 0, the mapping value is a prediction. The reciprocal of the value of the pixel in the data is twice, and the above 2 is larger than the value of the pixel in the prediction data, and N is an integer. When the value of a pixel in the prediction data is 0, the mapped value is a preset constant, typically 128 or 0.

S1004: Multiply the map value by the sum of the pixel values of the pixel points in the prediction data to obtain a fifth value.

S1005: Move the fifth value to the right by N bits to obtain the sixth value.

Since the mapping value is 2^ times the reciprocal of the value of the pixel in the prediction data, and multiplied by the mapping value of the value of the pixel in the prediction data, the obtained fifth value is directly compared to the pixel in the prediction data. The sum of the pixel values is divided by the value of the pixel by 2^ times. Moving a value to the right by N bits in binary is equivalent to reducing the value by 2^ times, shifting a value to the left by N bits. , which is equivalent to increasing the value by a multiple. Therefore, the sixth value can be obtained by shifting the first value to the N position by S1005, so that the sixth value is directly compared with the pixel value of the pixel point in the prediction data. The result of dividing the values of the points is equal. S1006: The sixth value is taken as the mean value. In this embodiment, the value of the pixel in the prediction data is correspondingly mapped by the lookup table, which avoids the complicated operation of the division operation. In this embodiment, the size of the lookup table is only the same as the pixel point in the prediction data. The value range of the value is related, so the implementation complexity is reduced to some extent, the chip area is reduced, and the processing speed is increased.

Further, after acquiring the values of the pixel points in the prediction data, the values of the pixel points in the prediction data are shifted to the right by M bits, so as to narrow the value range of the pixel values in the prediction data, and then pass the A numerical value shifting operation is performed such that the second value is equal to the result of directly dividing the sum of the pixel values of the pixel points in the predicted data by the value of the pixel point, thereby further reducing the complexity of the implementation. The method of this embodiment specifically includes :

S1101: Obtain prediction data of a current image block of the depth image.

S1102: Obtain a sum value of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data.

S1103: The value of the pixel in the prediction data is shifted to the right by M bits, and the value of the pixel in the shifted prediction data is obtained.

The value of the pixel in the prediction data is shifted to the right by M bits, which is equivalent to reducing the value of the pixel in the prediction data to the original 1/2 ^M .

SI 104: Obtain a mapping value corresponding to a value in the lookup table according to the value of the pixel point in the shifted prediction data.

The lookup table contains two sets of data, which are mapping values corresponding to the values of the pixel points in the shifted prediction data and the values of the pixel points in the shifted prediction data, and the values of the pixel points in the predicted data are not When 0, the mapping value is 2^ times the reciprocal of the numerical value of the pixel in the prediction data, and the above 2 is larger than the numerical value of the pixel in the prediction data, and N is an integer. When the number of pixels in the predicted data is 0, the mapped value is a preset constant, typically 128 or 0.

S1105: Multiply the map value by the sum of the pixel values of the pixel points in the prediction data to obtain a fifth value.

S1106: Move the fifth value to the right by N plus M bits to obtain the sixth value.

Since the value of the pixel in the prediction data is first shifted to the right by M bits, and then 2^ is divided by the value of the pixel in the shifted prediction data to obtain a mapping value, which is compared with the pixel in the prediction data. After multiplying the mapped values of the values, the obtained fifth value is increased by ^+M times than the result of directly dividing the sum of the pixel values of the pixel points in the predicted data with the values of the pixel points, and therefore, the S606 can be The fifth value is shifted right by N+M bits to obtain a sixth value such that the sixth value is equal to the result of directly dividing the sum of the pixel values of the pixel points in the prediction data by the values of the pixel points.

S1107: The sixth value is taken as the mean value.

In this embodiment, the value of the pixel in the prediction data is shifted to the right by M bits, and then the value of the pixel in the shifted prediction data is correspondingly mapped by the lookup table, thereby avoiding the division operation. A complicated problem is implemented, and the size of the lookup table in the embodiment is only related to the value range of the values of the pixel points in the shifted prediction data, thereby further reducing the complexity of the implementation and reducing the chip area. Improve processing speed.

To further reduce the complexity of the implementation, the method in this embodiment specifically includes:

S1201: Obtain prediction data of a current image block of the depth image.

S1202: Obtain a sum of pixel values of pixel points in the prediction data and a value of the pixel points in the prediction data.

S1203: Determine whether the value of the pixel in the predicted data is greater than a preset value. If yes, execute S1204-S1207, if no, execute S1208~S1210.

S1204: The value of the pixel in the prediction data is shifted to the right by M bits, and the value of the pixel in the shifted prediction data is obtained.

S1205: Obtain a mapping value corresponding to the value in the lookup table according to the value of the pixel in the shifted prediction data.

S1206: Multiply the map value by the sum of the pixel values of the pixel points in the prediction data to obtain a fifth value.

S1207: Move the fifth value to the right by N plus M bits to obtain the sixth value.

The above steps S1204~S1207 are similar to the corresponding steps in FIG. 11, and details are not described herein again.

S1208: Obtain a mapping value corresponding to the value in the lookup table according to the value of the pixel in the prediction data;

S1209: Multiply the mapping value by a sum of pixel values of pixel points in the prediction data to obtain a fifth value;

S1210: Move the fifth value to the right by N bits to obtain the sixth value.

The above steps S1208~S1210 are similar to the corresponding steps in FIG. 10, and details are not described herein again. S1211 : The sixth value is taken as the mean value.

In the above steps, the complexity of the implementation may be reduced to some extent due to the addition of the judgment condition. For example, when the preset value is 2 1 ^ΰ , when the pixel value of the pixel of the prediction data is 2 ⁸ , the search is performed according to 2 ⁸ The mapping value corresponding to ²⁸ is obtained in the table; the mapping value is multiplied by the sum of the pixel values of the pixel points in the prediction data to obtain a fifth value, and the fifth value is right-shifted to obtain a sixth value. When the pixel value of the pixel of the prediction data is 2 ¹² , the ^{12 12 is} shifted to the right by 4 bits, and after shifting to 2 ⁸ , the mapping value corresponding to ²⁸ is obtained according to 2 ⁸ in the lookup table; The sum value of the pixel values of the pixel points in the prediction data is multiplied to obtain a fifth value, and the fifth value is shifted right by Ν+Μ to obtain a sixth value. It can be seen that the values of the pixel points in the two different prediction data in the above process can be obtained by searching the same record in the lookup table. Therefore, the complexity of the implementation can be further reduced and the size of the lookup table can be reduced by the method of the present embodiment.

In this embodiment, after obtaining the value of the pixel point in the prediction data, adding a judgment condition, determining whether the value of the pixel point in the prediction data is greater than a preset value, when the value of the pixel point in the prediction data is greater than the pre-predetermined value When the value is set, the operation corresponding to the shift of the pixel points is used, the division operation is avoided, and the size of the lookup table is only related to the value range of the pixel values in the shifted prediction data, which can be further reduced. The complexity of the implementation, when the value of the pixel in the prediction data is less than the preset value, the accuracy of the mapping value that can be obtained by the operation corresponding to the pixel point value is not shifted; The complexity of the implementation is reduced. One of ordinary skill in the art will appreciate that all or a portion of the steps to implement the various method embodiments described above can be accomplished by hardware associated with the program instructions. The aforementioned program can be stored in a computer readable storage medium. The program, when executed, performs the steps including the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

It should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, and are not intended to be limiting; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art The technical solutions described in the foregoing embodiments may be modified, or some or all of the technical features may be equivalently replaced; and the modifications or substitutions do not deviate from the technical solutions of the embodiments of the present invention. range.

Claims

Claim

A method for encoding a depth image, comprising:

Obtaining prediction data of the current image block of the depth image, obtaining a predicted pixel value in the prediction data according to a preset step size, calculating a first mean value of the prediction data according to the predicted pixel value, the preset step size a positive integer that is not 1;

The residual of the current image block is encoded.

The method according to claim 1, wherein the obtaining the residual of the current image block according to the first mean value of the prediction data and the pixel value of the pixel point of the image block comprises:

Obtaining data of the current image block of the depth image, obtaining a predicted pixel value in the data of the current image block according to a preset step size, and calculating a data of the current image block according to the pixel value of the data of the current image block. And a difference between the second mean value of the data of the current image block and the first mean value of the reference data of the current image block to obtain a residual of the current image block.

The method according to claim 1, wherein the obtaining the predicted pixel value in the prediction data according to the preset step size comprises:

The method according to any one of claims 1 to 3, further comprising: encoding the value of the preset step size.

The method according to any one of claims 1 to 4, wherein the preset step size is used to identify an interval between predicted data position coordinates.

The method according to any one of claims 1 to 5, wherein the encoding the residual of the current image block comprises: Mapping the residual to a residual mapping value, indicating that the bit of the residual mapping value is smaller than a bit representing the residual;

The residual map value is encoded.

The method according to any one of claims 1 to 6, wherein before the obtaining the predicted pixel value in the prediction data according to the preset step size, the method further includes:

Determining, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set;

If so, it is determined that the predicted pixel value is obtained in the prediction data according to the preset step size.

8. The method according to claim 7, wherein the set of intra prediction modes comprises at least one of intra prediction modes described below:

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

9. A method for decoding a depth image, comprising:

Obtaining a residual of the current image block;

A pixel value of a pixel point of the current image block is obtained according to a first mean value of the prediction data and a residual of the current image block.

The method according to claim 9, wherein the obtaining the pixel value of the pixel of the current image block according to the first mean value of the prediction data and the residual of the current image block comprises: Adding a first mean value of the prediction data of the current image block to a residual of the current image block to obtain a pixel value of a pixel point of the current image block; or

The method according to claim 9, wherein the step is according to a preset step size Obtaining predicted pixel values in the forecast data, including:

The method according to any one of claims 9 to 11, further comprising: parsing the code stream to obtain a value identifying the step size.

The method according to any one of claims 9 to 12, wherein the preset step is used to identify an interval between predicted data position coordinates.

The method according to any one of claims 9 to 13, wherein the obtaining the residual of the current image block comprises:

The method according to any one of claims 9 to 14, wherein before the obtaining the predicted pixel value in the prediction data according to the preset step size, the method further includes:

16. The method according to claim 15, wherein the set of intra prediction modes comprises at least one of the following intra prediction modes:

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

17. An apparatus for encoding a depth image, comprising: a first processing module, configured to obtain prediction data of a current image block of the depth image, obtain a predicted pixel value in the prediction data according to a preset step size, and calculate a first mean value of the prediction data according to the predicted pixel value, The preset step size is a positive integer that is not 1;

The encoding device according to claim 17, wherein the second processing module is specifically configured to compare a pixel point of the current image block with a first mean value of prediction data of the current image block, Obtaining a residual of the current image block; or obtaining data of a current image block of the depth image, obtaining a predicted pixel value in the data of the current image block according to a preset step size, according to data of the current image block Predicting a pixel value to calculate a second mean value of the data of the current image block, and comparing a second mean value of the data of the current image block with a first mean value of the predicted data of the current image block to obtain the current image block Residual.

The encoding device according to claim 17, wherein the first processing module is configured to obtain a predicted pixel value according to an equal step size of a horizontal direction and a vertical direction according to a size of the current image block; or Obtaining a predicted pixel value according to the unequal step size of the horizontal direction and the vertical direction according to the size of the current image block; or determining the preset step size according to the size of the current image block, according to the preset The step size gets the predicted pixel value.

The encoding device according to any one of claims 17 to 19, wherein the encoding module is further configured to encode the value of the preset step size.

The encoding apparatus according to any one of claims 17 to 20, wherein the preset step size is used to identify an interval between predicted data position coordinates.

The coding apparatus according to any one of claims 17 to 21, wherein the coding module is specifically configured to map the residual to a residual mapping value, and to represent the residual mapping value. The bit is smaller than the bit representing the residual, and the residual map value is encoded.

The encoding device according to any one of claims 17 to 22, wherein the first processing module is further configured to determine, according to an intra prediction mode of a current image block, whether the intra prediction mode belongs to Presetting a set of intra prediction modes, if yes, determining to obtain the predicted pixel value in the prediction data according to the preset step size.

The encoding apparatus according to claim 23, wherein the intra prediction mode set The method includes at least one of the following intra prediction modes:

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

25. A decoding apparatus for a depth image, comprising:

a processing module, configured to obtain prediction data of a current image block of the depth image, obtain a predicted pixel value in the prediction data according to a preset step size, and calculate a first mean value of the prediction data according to the predicted pixel value, The preset step size is a positive integer that is not 1;

The decoding device according to claim 25, wherein the decoding module is configured to add a first mean value of the prediction data of the current image block and a residual of the current image block to obtain a a pixel value of a pixel of the current image block; or, adding a first mean value of the prediction data of the current image block to the residual, obtaining a pixel mean of the current image block, and using the current image block The pixel mean is the pixel value of the pixel of the current image block.

The decoding device according to claim 25, wherein the processing module is configured to obtain a predicted pixel value according to an equal step size of a horizontal direction and a vertical direction according to a size of the current image block; or The size of the current image block is obtained by unequal steps in the horizontal direction and the vertical direction to obtain a predicted pixel value; or, determining the preset step size according to the size of the current image block, according to the preset step The predicted pixel value is obtained long.

The decoding apparatus according to any one of claims 25 to 27, wherein the decoding module is further configured to parse the code stream to obtain a value identifying the step size.

The decoding apparatus according to any one of claims 25 to 28, wherein the preset step size is used to identify an interval between predicted data position coordinates.

The decoding device according to any one of claims 25 to 29, wherein the decoding module is configured to perform decoding processing on the encoding result of the current image block to obtain a residual of the current image block. a difference mapping value, mapping the residual mapping value to a residual, using the residual as a residual of the current image block, indicating that a bit of the residual mapping value is smaller than a bit representing the residual .

The decoding device according to any one of claims 25 to 30, wherein the processing The module is further configured to determine, according to an intra prediction mode of the current image block, whether the intra prediction mode belongs to a preset intra prediction mode set, and if so, determine, according to the preset step size, in the prediction data. The predicted pixel value is obtained.

The decoding apparatus according to claim 31, wherein the intra prediction mode set includes at least one of intra prediction modes described below:

DC DC mode;

Planar Planar mode;

Explicitly identified Wedgelet Wedgelet mode;

Wedgelet Wedgelet mode based on intra prediction mode.

33. A method for encoding a depth image, comprising:

Obtaining prediction data of the current image block of the depth image, performing arithmetic operations on each N pixel points in the pixel in the prediction data, obtaining an arithmetic value of the corresponding pixel value, and calculating the prediction data according to the arithmetic value of the corresponding pixel value First mean value;

The residual of the current image block is encoded.

The method according to claim 33, wherein the performing arithmetic operations on each of the pixel points in the pixel in the prediction data to obtain an arithmetic value of the corresponding pixel value includes:

An averaging operation is performed on every N pixel points in the pixel in the prediction data to obtain an average pixel value, and the average pixel value is used as an arithmetic value of the corresponding pixel.

The method according to claim 33 or 34, wherein the residual is obtained according to the first mean value of the prediction data and the pixel value of the pixel of the image block, including:

Placing a pixel of the image block with a first mean of the prediction data of the current image block to obtain a residual of the current image block;

36. A method for decoding a depth image, comprising:

Obtaining prediction data of the current image block of the depth image, for every N pixels in the pixel in the prediction data Performing an arithmetic operation to obtain an arithmetic value of a corresponding pixel value, and calculating a first mean value of the predicted data according to an arithmetic value of the corresponding pixel value;

Obtaining a residual of the current image block;

The method according to claim 36, wherein the performing an arithmetic operation on each of the pixel points in the pixel in the prediction data to obtain an arithmetic value of the corresponding pixel value includes:

An average operation is performed on each of the N pixels in the current image block and the predicted data to obtain an average pixel value, and the average pixel value is used as an arithmetic value of the corresponding pixel value.

The method according to claim 36 or 37, wherein: obtaining a pixel value of a pixel of the current image block according to an average of the prediction data and a residual of the current image block, including: Adding a mean value of the prediction data of the current image block to a residual of the current image block to obtain a pixel value of a pixel point of the current image block;

39. An apparatus for encoding a depth image, comprising:

a first processing module, configured to obtain prediction data of a current image block of the depth image, perform an arithmetic operation on each N pixel points in the pixel in the prediction data, obtain an arithmetic value of the corresponding pixel value, and perform arithmetic according to the corresponding pixel value Calculating a first mean value of the predicted data;

The encoding device according to claim 39, wherein the first processing module is configured to perform an averaging operation on every N pixel points in the pixel in the prediction data to obtain an average pixel value, The average pixel value is taken as an arithmetic value of the corresponding pixel.

The encoding device according to claim 39 or 40, wherein the second processing module is specifically configured to compare a pixel point of the image block with a first mean value of prediction data of the current image block. Obtaining a residual of the current image block; or obtaining data of a current image block of the depth image, obtaining the pixel value in data of the current image block according to a preset step size, and calculating according to the pixel value a second mean value of the data of the image block is obtained by comparing a second mean value of the image block with a first mean value of a reference image block of the current image block to obtain a residual of the current image block.

42. A decoding apparatus for a depth image, comprising:

a processing module, configured to obtain prediction data of a current image block of the depth image, perform an arithmetic operation on each N pixel points in the pixel in the prediction data, obtain an arithmetic value of the corresponding pixel value, and calculate according to an arithmetic value of the corresponding pixel value a first mean of the predicted data;

The decoding device according to claim 42, wherein the processing module is configured to perform an averaging operation on each of the N pixels in the current image block and the prediction data to obtain an average pixel value. The average pixel value is taken as an arithmetic value of the corresponding pixel value.

The decoding device according to claim 42 or 43, wherein the decoding module is specifically configured to add an average value of prediction data of the current image block to a residual of the current image block to obtain a a pixel value of a pixel of the current image block; or, adding a mean value of the prediction data of the current image block to the residual, obtaining a pixel mean of the current image block, and pixels of the current image block The mean value is the pixel value of the pixel point of the current image block.