CN104159121A

CN104159121A - Method and device for coding and decoding image block based on depth information

Info

Publication number: CN104159121A
Application number: CN201410299473.6A
Authority: CN
Inventors: 朱策; 张玉花; 林永兵; 郑建铧; 宋剑军
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2014-06-26
Filing date: 2014-06-26
Publication date: 2014-11-19

Abstract

The invention discloses a method and a device for coding and decoding an image block based on depth information. According to a preset partitioned mode having two partitions or according to a partitioned mode corresponding to a comparison result of mask values of two pixel points, movement compensation prediction of the image block is carried out, and then prediction information of the image block is obtained, thereby coding or decoding the image block. An embodiment of the invention is provided with a corresponding device. According to the coding and decoding method provided by an embodiment of the invention, in the coding and decoding, the preset partitioned mode is directly used or a partitioned mode is determined by the comparison result of the mask values, so that the partitioned mode can be quickly determined, and the coding and decoding efficiency of the image block based on DBBP (Depth-based Block Partitioning) is improved.

Description

A kind of coding of the image block based on depth information, coding/decoding method and device

Technical field

The present invention relates to coding and decoding technical field, be specifically related to a kind of coding, coding/decoding method and device of the image block based on depth information.

Background technology

Piece subregion DBBP based on the degree of depth (Depth-based Block Partitioning) Predicting Technique is to utilize the parallax information of current texture block to be predicted, in the depth map of having encoded, based on base viewpoint, find corresponding virtual depth piece, the information of the virtual depth piece that utilization is found, texture block is divided into prospect and two regions of background, and be labeled as respectively different mask (mask) value, and obtain by series of computation the partitioned mode that current texture piece is comparatively suitable, then according to partitioned mode, texture block is carried out to two prediction signal that motion compensation obtains texture block, according to the mask information of current texture piece, two prediction signal that obtain are merged and form the final prediction signal of current texture piece.

The process that obtains by series of computation the partitioned mode that current texture piece is comparatively suitable in prior art is:

While take 4 down-sampling that is coefficient to carry out as example, first virtual depth piece is divided into the fritter of 4*4, the angle pixel value of the top left pixel point of each fritter is added and gets average, according to the average of calculating, the pixel value that travels through the pixel in upper left of each 4*4 fritter carries out matched filtering, from Nx2N and two kinds of partitioned modes of 2NxN, select the comparatively partitioned mode of coupling.Specific algorithm is as follows:

Variable declaration:

T: the pixel value of the top left corner pixel point of each 4*4 virtual depth fritter is added the average obtaining after summation;

Segment: the magnitude relationship that records the average T of current virtual depth pixel value and calculating;

HalfSize: the half-size scale of current texture block size (long and wide equating);

MatchedPartSum[2] [2]: for the state of record matching filtering;

Travel through successively the pixel value of the top left corner pixel point of each 4*4 virtual depth fritter, suppose the current pixel (x that traverses, y), according to the magnitude relationship of the pixel value of (x, y) and T, obtain pixel (x, y) Segment value is mated, and if the abscissa x of the pixel (x, y) traversing current is less than HalfSize, matchedPartSum[0] [Segment] ++, otherwise matchedPartSum[0] [1-Segment] ++; If the ordinate y of the pixel (x, y) traversing current is less than HalfSize, matchedPartSum[1] [Segment] ++, otherwise matchedPartSum[1] [1-Segment] ++

Big or small, from matchedPartSum[0] [Segment], matchedPartSum[0] [1-Segment], matchedPartSum[1] [Segment], matchedPartSum[1] select maximum [1-Segment], if maximum is matchedPartSum[0] [Segment] or matchedPartSum[0] [1-Segment] partitioned mode be Nx2N, otherwise partitioned mode is 2NxN.

As can be seen here, in prior art, when determining partitioned mode, to repeatedly carry out addition, division and the operation such as relatively, cause determining that the complexity of partitioned mode is too high, thereby reduced the efficiency of the encoding and decoding based on DBBP image block.

Summary of the invention

The embodiment of the present invention is for solving in prior art in the encoding-decoding process based on DBBP, owing to determining that the complexity of partitioned mode is high, cause the problem that encoding-decoding efficiency is low, the present invention proposes a kind of coding, coding/decoding method of the image block based on depth information, can determine fast partitioned mode, thereby improve the efficiency of the encoding and decoding based on DBBP image block.The embodiment of the present invention also provides corresponding device.

First aspect present invention provides a kind of coding method of the image block based on depth information, comprising:

Obtain current image block to be encoded;

According to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded;

According to the motion vector information of described the first subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal;

According to the motion vector information of described the second subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal;

According to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

According to the prediction signal of each pixel described in described current image block to be encoded, the coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

In conjunction with first aspect, in the possible implementation of the first, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with first aspect or the possible implementation of first aspect the first, in the possible implementation of the second, described mask value is 0 or 1,

Describedly according to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, comprising:

When the mask value of described each pixel is 0, select described the first prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the second prediction signal as the prediction signal of described each pixel; Or,

When the mask value of described each pixel is 0, select described the second prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the first prediction signal as the prediction signal of described each pixel.

Second aspect present invention provides a kind of coding/decoding method of the image block based on depth information, comprising:

From code stream, obtain the prediction sign of current image block to be decoded, described prediction sign is used to indicate the prediction that described current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth;

According to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be decoded;

According to the motion vector information of described the first subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal;

According to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal;

According to the mask value of each pixel in described current image block to be decoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

Prediction signal according to each pixel described in described current image block to be decoded, obtains decoded image block.

In conjunction with second aspect, in the possible implementation of the first, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with second aspect or the possible implementation of second aspect the first, in the possible implementation of the first, described mask value is 0 or 1,

Third aspect present invention provides a kind of coding method of the image block based on depth information, comprising:

Obtain current image block to be encoded, and according to the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion;

According to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels;

According to the corresponding partitioned mode of the comparative result of the mask value of described two pixels, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded;

In conjunction with the third aspect, in the possible implementation of the first, during the described mask value according to two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion, described method also comprises:

Obtain at least one in the mask value of two other pixel except described two pixels in the pixel at described four angles;

Described according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels, comprising:

According to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of combination of the first comparative result and the second comparative result;

Wherein, the comparative result that described the first comparative result is the mask value of described two pixels;

Described the second comparative result be in the mask value of a pixel in described two pixels and described two other pixel with the comparative result of the mask value of the pixel of a described pixel on same straight line.

In conjunction with the third aspect or the possible implementation of third aspect the first, in the possible implementation of the second, described partitioned mode in the corresponding relation of described preset comparative result and partitioned mode comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with the third aspect, third aspect the first or the possible implementation of the second, in the third possible implementation, described mask value is 0 or 1,

Fourth aspect present invention provides a kind of coding/decoding method of the image block based on depth information, comprising:

From code stream, obtain current image block to be decoded prediction sign, described prediction sign is used to indicate the prediction that current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth;

According to depth information, obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be decoded;

According to the mask value of each pixel in described current image block to be decoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel;

In conjunction with fourth aspect, in the possible implementation of the first, during the described mask value according to two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion, described method also comprises:

In conjunction with fourth aspect or the possible implementation of fourth aspect the first, in the possible implementation of the second, described partitioned mode in the corresponding relation of described preset comparative result and partitioned mode comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with fourth aspect, fourth aspect the first or the possible implementation of the second, in the third possible implementation, described mask value is 0 or 1,

Fifth aspect present invention provides a kind of code device, comprising:

Acquiring unit, for obtaining current image block to be encoded;

Obtain unit, for according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of the image block current to be encoded that described acquiring unit obtains;

Motion compensated prediction unit, be used for according to the motion vector information of described first subregion of described acquisition unit acquisition, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described acquisition unit, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell, be used for according to the mask value of described current each pixel of image block to be encoded, described the first prediction signal obtaining from the prediction of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

Coding unit after prediction, prediction signal for each pixel described in the described current image block to be encoded obtaining according to described selected cell choosing, coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

In conjunction with the 5th aspect, in the possible implementation of the first, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with the 5th aspect or the possible implementation of the 5th aspect the first, in the possible implementation of the second, described mask value is 0 or 1,

Described selected cell specifically for:

Sixth aspect present invention provides a kind of decoding device, comprising:

Acquiring unit, for obtain the prediction sign of current image block to be decoded from code stream, described prediction sign is used to indicate the prediction that described current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth;

Obtain unit, for according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of the described current image block to be decoded that described acquiring unit obtains;

Motion compensated prediction unit, be used for according to the motion vector information of described first subregion of described acquisition unit acquisition, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described acquisition unit, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell, be used for according to the mask value of described current each pixel of image block to be decoded, described the first prediction signal obtaining from the prediction of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

Decoding unit after prediction, the prediction signal for each pixel described in the described current image block to be decoded of selecting to obtain according to described selected cell, obtains decoded image block.

In conjunction with the 6th aspect, in the possible implementation of the first, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with the 6th aspect or the possible implementation of the 6th aspect the first, in the possible implementation of the second, described mask value is 0 or 1,

Described selected cell specifically for:

Seventh aspect present invention provides a kind of code device, comprising:

Acquiring unit, for obtaining current image block to be encoded, and according to the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion;

First obtains unit, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of comparative result of the mask value of described two pixels that described acquiring unit obtains;

Second obtains unit, for according to the corresponding partitioned mode of comparative result of the mask value of described two pixels of described the first acquisition unit acquisition, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded;

Motion compensated prediction unit, be used for according to the motion vector information of described first subregion of described the second acquisition unit acquisition, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described the second acquisition unit, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Coding unit after prediction, prediction signal for each pixel described in the described current image block to be encoded of selecting to obtain according to described selected cell, coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

In conjunction with the 7th aspect, in the possible implementation of the first,

Described acquiring unit, also for obtaining at least one in the mask value of two other pixel of the pixel at described four angles except described two pixels;

Described first obtains unit, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of combination of the first comparative result and the second comparative result;

In conjunction with the 7th aspect or the possible implementation of the 7th aspect the first, in the possible implementation of the second, described preset comparative result and partitioned mode comprise two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with the 7th aspect, the 7th aspect the first or the possible implementation of the second, in the possible implementation of the second, described mask value is 0 or 1,

Described selected cell specifically for:

Eighth aspect present invention provides a kind of decoding device, comprising:

Acquiring unit, for from code stream, obtain current image block to be decoded prediction sign, described prediction sign is used to indicate the prediction that current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth, and according to depth information, obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be decoded; First obtains unit, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of comparative result of the mask value of described two pixels that described acquiring unit obtains;

Motion compensated prediction unit, be used for according to the motion vector information of described first subregion of described the second acquisition unit acquisition, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal, according to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell, be used for according to the mask value of described current each pixel of image block to be decoded, described the first prediction signal obtaining from the prediction of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel;

In conjunction with eight aspect, in the possible implementation of the first,

In conjunction with eight aspect or the possible implementation of eight aspect the first, in the possible implementation of the second,

Described partitioned mode in the corresponding relation of described preset comparative result and partitioned mode comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In conjunction with eight aspect, eight aspect the first or the possible implementation of the second, in the possible implementation of the second,

Described mask value is 0 or 1,

Described selected cell specifically for:

With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the method of the encoding and decoding that the embodiment of the present invention provides, when encoding and decoding, directly use preset partitioned mode or determine that by the comparative result of mask value partitioned mode can determine partitioned mode fast, thereby improved the efficiency of the encoding and decoding based on DBBP image block.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Figure 1A is an embodiment schematic diagram of the coding method of the image block based on depth information in the embodiment of the present invention;

One example schematic diagram of the partitioned mode of the image block based on depth information in Figure 1B embodiment of the present invention;

Another example schematic diagram of the partitioned mode of the image block based on depth information in Fig. 1 C embodiment of the present invention;

Fig. 2 is an embodiment schematic diagram of the coding/decoding method of the image block based on depth information in the embodiment of the present invention;

Fig. 3 is an embodiment schematic diagram of the coding method of the image block based on depth information in the embodiment of the present invention;

Fig. 4 is an example schematic diagram of the coding/decoding method of the image block in the embodiment of the present invention;

Fig. 5 is an embodiment schematic diagram of the coding/decoding method of the image block based on depth information in the embodiment of the present invention;

Fig. 6 is an embodiment schematic diagram of code device in the embodiment of the present invention;

Fig. 7 is an embodiment schematic diagram of decoding device in the embodiment of the present invention;

Fig. 8 is an embodiment schematic diagram of code device in the embodiment of the present invention;

Fig. 9 is an embodiment schematic diagram of decoding device in the embodiment of the present invention;

Figure 10 is an embodiment schematic diagram of code device in the embodiment of the present invention;

Figure 11 is an embodiment schematic diagram of decoding device in the embodiment of the present invention;

Figure 12 is an embodiment schematic diagram of code device in the embodiment of the present invention;

Figure 13 is an embodiment schematic diagram of decoding device in the embodiment of the present invention.

Embodiment

The embodiment of the present invention provides the embodiment of the present invention to be provided as to solve in prior art in the encoding-decoding process based on DBBP, owing to determining that the complexity of partitioned mode is high, cause the problem that encoding-decoding efficiency is low, the present invention proposes a kind of coding, coding/decoding method of the image block based on depth information, can determine fast partitioned mode, thereby improve the efficiency of the encoding and decoding based on DBBP image block.The embodiment of the present invention also provides corresponding device.Below be elaborated respectively.

In order to make those skilled in the art person understand better the present invention program, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the embodiment of a part of the present invention, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, should belong to the scope of protection of the invention.

Consult Figure 1A, an embodiment of the coding method of a kind of image block based on depth information that the embodiment of the present invention provides comprises:

101, obtain current image block to be encoded.

102,, according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded.

The described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

In the embodiment of the present invention, the partitioned mode of image block can comprise the partitioned modes such as Nx2N, 2NxN, 2NxnU, 2NxnD, nLx2N or nRx2N.

The partitioned mode of 2NxN can be understood as image block is divided into two from horizontal direction, above sub-block account for 1/2, below sub-block account for 1/2; Nx2N can be understood as image block is divided into two from vertical direction, and left side sub-block accounts for 1/2, and the right accounts for 1/2.Partitioned mode is as shown in Figure 1B the partitioned mode of 2NxN, the partitioned mode that partitioned mode is as shown in Figure 1 C Nx2N.

The partitioned mode of 2NxnU can be understood as texture maps is divided into two from horizontal direction, above sub-block account for 1/4, below sub-block account for 3/4; The partitioned mode of 2NxnD can be understood as texture maps is divided into two from horizontal direction, above sub-block account for 3/4, below sub-block account for 1/4; NLx2N can be understood as image block is divided into two from vertical direction, and left side sub-block accounts for 1/4, and the right accounts for 3/4; NRx2N can be understood as image block is divided into two from vertical direction, and left side sub-block accounts for 3/4, and the right accounts for 1/4.

103, according to the motion vector information of described the first subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal.

104, according to the motion vector information of described the second subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal.

105, according to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion.

The mask value of each pixel can obtain according to the virtual depth piece of current image block to be encoded, calculate the mean value of the pixel value of each pixel in virtual depth piece, in virtual depth piece, the pixel value of each pixel and this mean value compare, when the pixel value of pixel is greater than this mean value, the mask value of pixel in the image block front to be encoded of correspondence is labeled as to 0, be less than and be labeled as 1, finally obtain front image block to be encoded in the mask value of each pixel, when mask value is 0, select the prediction signal of the pixel that the first prediction signal is 0 as mask value, when mask value is 1, select the prediction signal of the pixel that the second prediction signal is 0 as mask value.Also can select conversely.Namely, when the mask value of described each pixel is 0, select described the second prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the first prediction signal as the prediction signal of described each pixel.Concrete selection mode is determined according to predefined.

106,, according to the prediction signal of each pixel described in described current image block to be encoded, the coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

The embodiment of the present invention is obtained current image block to be encoded; According to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded; According to the motion vector information of described the first subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal; According to the motion vector information of described the second subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal; According to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion; According to the prediction signal of each pixel described in described current image block to be encoded, the coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the method of the coding that the embodiment of the present invention provides, when coding, directly use preset partitioned mode, can determine fast partitioned mode, thereby improve the efficiency of the coding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 1, in the optional embodiment of the coding method of the image block based on depth information that the embodiment of the present invention provides, described mask value is 0 or 1,

Consult Fig. 2, an embodiment of the coding/decoding method of the image block based on depth information that the embodiment of the present invention provides comprises:

201, from code stream, obtain the prediction sign of current image block to be decoded, described prediction sign is used to indicate the prediction that described current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth.

202,, according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be decoded.

The partitioned mode of Nx2N can be understood as image block is divided into two from horizontal direction, above sub-block account for 1/2, below sub-block account for 1/2; 2NxN can be understood as image block is divided into two from vertical direction, and left side sub-block accounts for 1/2, and the right accounts for 1/2.

203, according to the motion vector information of described the first subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal.

204, according to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal.

205, according to the mask value of each pixel in described current image block to be decoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion.

The mask value of each pixel can obtain according to the virtual depth piece of current image block to be encoded, calculate the mean value of the pixel value of each pixel in virtual depth piece, in virtual depth piece, the pixel value of each pixel and this mean value compare, when the pixel value of pixel is greater than this mean value, the mask value of pixel in the image block front to be encoded of correspondence is labeled as to 0, be less than and be labeled as 1, finally obtain front image block to be encoded in the mask value of each pixel, when mask value is 0, select the prediction signal of the pixel that the first prediction signal is 0 as mask value, when mask value is 1, select the prediction signal of the pixel that the second prediction signal is 0 as mask value.Also can select conversely.Namely, when the mask value of described each pixel is 0, select described the second prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the first prediction signal as the prediction signal of described each pixel.

Concrete selection mode, depend on the partitioned mode of image block and the mask value that obtains based on depth information between relation.

206, according to the prediction signal of each pixel described in described current image block to be decoded, obtain decoded image block.

In the embodiment of the present invention, from the code stream of current image block to be decoded, obtain the prediction sign of described code stream, described prediction sign is used to indicate described code stream and has adopted the prediction of the piece subregion DBBP based on the degree of depth when coding, and indicates described current image block to be decoded to have two subregions; According to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be decoded; According to the motion vector information of described the first subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal; According to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal; According to the mask value of each pixel in described current image block to be decoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion; According to the prediction signal of each pixel described in described current image block to be decoded, the decoding after the code stream of described current image block to be decoded is predicted, obtains decoded image block.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the method of the decoding that the embodiment of the present invention provides, when decoding, directly use preset partitioned mode, can determine fast partitioned mode, thereby improve the efficiency of the decoding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 2, in another embodiment of the coding/decoding method of the image block based on depth information that the embodiment of the present invention provides, described mask value is 0 or 1,

Describedly according to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, can comprise:

Consult Fig. 3, an embodiment of the coding method of the image block based on depth information that the embodiment of the present invention provides comprises:

301, obtain current image block to be encoded, and according to the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion.

In the embodiment of the present invention, as shown in Figure 4, the texture block of 8*8 of take is example, texture block shown in Fig. 4 has 64 pixels, the pixel at four angles marks by 1,2,3,4 in Fig. 4, and wherein the mask value of pixel 1 represents with Mask1, and the mask value of pixel 2 represents with Mask2, the mask value of pixel 3 represents with Mask3, and the mask value of pixel 4 represents with Mask4.

Can only obtain Mask1 and Mask2, also can only obtain Mask1 and Mask3, also can only obtain Mask3 and Mask4, can also only obtain Mask2 and Mask4.

302,, according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels.

Mask1 and Mask2, Mask1 and Mask3, Mask3 and Mask4 or Mask2 and Mask4 are compared, obtain this two comparative results whether mask value is equal, corresponding 2NxN when preset comparative result can equate for the comparative result of two mask values on same horizontal line with the corresponding relation of partitioned mode, corresponding Nx2N when unequal, corresponding Nx2N when two mask values on same vertical curve equate, corresponding 2NxN when unequal.

For example: when Mask1 and Mask2 equate, definite partitioned mode is 2NxN, is Nx2N when unequal, when Mask1 and Mask3 equate, definite partitioned mode is Nx2N, and when unequal, definite partitioned mode is 2NxN.

303,, according to the corresponding partitioned mode of the comparative result of the mask value of described two pixels, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded.

304, according to the motion vector information of described the first subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal.

305, according to the motion vector information of described the second subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal.

306, according to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion.

307,, according to the prediction signal of each pixel described in described current image block to be encoded, the coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

In the embodiment of the present invention, obtain current image block to be encoded, and obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in described current image block to be encoded, according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels, according to the corresponding partitioned mode of the comparative result of the mask value of described two pixels, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded, according to the motion vector information of described the first subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, according to the motion vector information of described the second subregion, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal, according to the mask value of each pixel in described current image block to be encoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, described mask value arrives according to described Depth Information Acquistion, according to the prediction signal of each pixel described in described current image block to be encoded, coding after described current image block to be encoded is predicted, obtain the encoding code stream of described current image block to be encoded.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the method of the coding that the embodiment of the present invention provides, in when coding, by the comparative result of mask value, determine that partitioned mode can determine partitioned mode fast, thereby improved the efficiency of the coding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 3, in another embodiment of the coding method of the image block based on depth information that the embodiment of the present invention provides, during the described mask value according to two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion, described method also comprises:

In the embodiment of the present invention, can obtain four kinds of comparative results, be respectively: the first is that two horizontal mask values equate, two vertical mask values are unequal, the second is that two horizontal mask values equate, two vertical mask values are also equal, and the third is that two horizontal mask values are unequal, and two vertical mask values equate, the 4th kind is that two horizontal mask values are unequal, and two vertical mask values are also unequal.Can partitioned mode corresponding to preset the first comparative result be 2NxN, the partitioned mode that other three kinds of comparative results are corresponding be all Nx2N.

In the embodiment of the present invention, consult the example shown in Fig. 4, relatively after Mask1 and Mask2, obtain Mask1 and equate with Mask2, further compare Mask1 and whether equate with Mask3 or Mask4, take Mask3 as example, if Mask1 and Mask3 are unequal, can obtain the partitioned mode of the associated 2NxN of this result, if Mask1 equate with Mask3, can obtain the partitioned mode of the associated Nx2N of this result.If Mask1 and Mask2 are unequal, no matter whether Mask1 equates with Mask3, all can obtain the partitioned mode of Nx2N.

Conversely, if Mask3 equates with Mask4, whether comparison Mask3 equates with Mask1 or Mask2, take Mask1 as example, if Mask3 and Mask1 are unequal, can obtain the partitioned mode of the associated 2NxN of this result, if Mask3 equate with Mask1, can obtain the partitioned mode of the associated Nx2N of this result.If Mask3 and Mask4 are unequal, no matter whether Mask3 equates with Mask1, all can obtain the partitioned mode of Nx2N.

Certainly, in the present embodiment, just illustrate, every kind of corresponding partitioned mode of result is unfixed, and with preset relevant, which kind of preset partitioned mode corresponding to result be, will obtain corresponding partitioned mode according to this result.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 3, in another embodiment of the coding method of the image block based on depth information that the embodiment of the present invention provides, described mask value is 0 or 1,

Consult Fig. 5, an embodiment of the coding/decoding method of the image block based on depth information that the embodiment of the present invention provides comprises:

401, from code stream, obtain current image block to be decoded prediction sign, described prediction sign is used to indicate the prediction that current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth, according to depth information, obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be decoded.

402,, according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels.

403,, according to the corresponding partitioned mode of the comparative result of the mask value of described two pixels, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded.

404, according to the motion vector information of described the first subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal.

405, according to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal.

406, according to the mask value of each pixel in described current image block to be decoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel.

407,, according to the prediction signal of each pixel described in described current image block to be decoded, the decoding after the code stream of described current image block to be decoded is predicted, obtains decoded image block.

In the embodiment of the present invention, from code stream, obtain current image block to be decoded prediction sign, described prediction sign is used to indicate the prediction that current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth, according to depth information, obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be decoded, according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels; According to the corresponding partitioned mode of the comparative result of the mask value of described two pixels, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded; According to the motion vector information of described the first subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal; According to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal; According to the mask value of each pixel in described current image block to be decoded, from described the first prediction signal and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel; According to the prediction signal of each pixel described in described current image block to be decoded, the decoding after the code stream of described current image block to be decoded is predicted, obtains decoded image block.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the method of the decoding that the embodiment of the present invention provides, in when decoding, by the comparative result of mask value, determine that partitioned mode can determine partitioned mode fast, thereby improved the efficiency of the decoding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 5, in another embodiment of the coding/decoding method of the image block based on depth information that the embodiment of the present invention provides, during the mask value of two pixels in the described pixel that obtains four angles in described current image block to be encoded on same horizontal line or same vertical curve, described method can also comprise:

Described according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of comparative result of the mask value of described two pixels, can comprise:

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 5, in another embodiment of the coding/decoding method of the image block based on depth information that the embodiment of the present invention provides, described mask value is 0 or 1,

Consult Fig. 6, an embodiment of the code device 50 that the embodiment of the present invention provides comprises:

Acquiring unit 501, for obtaining current image block to be encoded;

Obtain unit 502, for according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of the image block current to be encoded that described acquiring unit 501 obtains;

Motion compensated prediction unit 503, be used for according to the motion vector information of described first subregion of described acquisition unit 502 acquisitions, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described acquisition unit, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell 504, be used for according to the mask value of described current each pixel of image block to be encoded, described the first prediction signal obtaining from 503 predictions of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

Coding unit 505 after prediction, prediction signal for each pixel described in the described current image block to be encoded obtaining according to described selected cell choosing, coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

In the embodiment of the present invention, acquiring unit 501 obtains current image block to be encoded, obtain unit 502 according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of the image block current to be encoded that described acquiring unit 501 obtains, the motion vector information of described the first subregion that motion compensated prediction unit 503 obtains according to described acquisition unit 502, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described acquisition unit, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal, selected cell 504 is according to the mask value of each pixel in described current image block to be encoded, described the first prediction signal obtaining from 503 predictions of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, , described mask value arrives according to described Depth Information Acquistion, the prediction signal of each pixel described in the described current image block to be encoded that after prediction, coding unit 505 obtains according to described selected cell choosing, coding after described current image block to be encoded is predicted, obtain the encoding code stream of described current image block to be encoded.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the code device that the embodiment of the present invention provides, when coding, directly use preset partitioned mode, can determine fast partitioned mode, thereby improve the efficiency of the coding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 6, in another embodiment of the code device 50 that the embodiment of the present invention provides, described mask value is 0 or 1,

Described selected cell 504 specifically for:

Consult Fig. 7, an embodiment of the decoding device 60 that the embodiment of the present invention provides comprises:

Acquiring unit 601, for obtain the prediction sign of current image block to be decoded from code stream, described prediction sign is used to indicate the prediction that described current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth;

Obtain unit 602, for according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of the described current image block to be decoded that described acquiring unit 601 obtains;

Motion compensated prediction unit 603, be used for according to the motion vector information of described first subregion of described acquisition unit 602 acquisitions, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described acquisition unit 602, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell 604, be used for according to the mask value of described current each pixel of image block to be decoded, described the first prediction signal obtaining from the prediction of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

Decoding unit 605 after prediction, prediction signal for each pixel described in the described current image block to be decoded of selecting to obtain according to described selected cell 604, decoding after the code stream of described current image block to be decoded is predicted, obtains decoded image block.

In the embodiment of the present invention, acquiring unit 601 obtains the prediction sign of described code stream from the code stream of current image block to be decoded, described prediction sign is used to indicate described code stream and has adopted the prediction of the piece subregion DBBP based on the degree of depth when coding, and indicates described current image block to be decoded to have two subregions; Obtain unit 602 according to the preset partitioned mode that comprises two subregions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of the described current image block to be decoded that described acquiring unit 601 obtains; The motion vector information of described the first subregion that motion compensated prediction unit 603 obtains according to described acquisition unit 602, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described acquisition unit 602, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal; Selected cell 604 is according to the mask value of each pixel in described current image block to be decoded, described the first prediction signal obtaining from the prediction of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion; The prediction signal of each pixel described in the described current image block to be decoded that after prediction, decoding unit 605 is selected to obtain according to described selected cell 604, decoding after the code stream of described current image block to be decoded is predicted, obtains decoded image block.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the decoding device that the embodiment of the present invention provides, when decoding, directly use preset partitioned mode, can determine fast partitioned mode, thereby improve the efficiency of the decoding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 7, in another embodiment of the decoding device 60 that the embodiment of the present invention provides, described mask value is 0 or 1,

Described selected cell 604 specifically for:

Consult Fig. 8, an embodiment of the code device 70 that the embodiment of the present invention provides comprises:

Acquiring unit 701, for obtaining current image block to be encoded, and according to the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion; First obtains unit 702, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of comparative result of the mask value of described two pixels that described acquiring unit 701 obtains;

Described preset comparative result and partitioned mode comprise two subregions that are arranged above and below, or two subregions of left and right arrangement.

Second obtains unit 703, for according to the corresponding partitioned mode of comparative result of the mask value of described two pixels of described the first acquisition unit 702 acquisitions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded;

Motion compensated prediction unit 704, be used for according to the motion vector information of described first subregion of described the second acquisition unit 703 acquisitions, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described the second acquisition unit, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell 705, be used for according to the mask value of described current each pixel of image block to be encoded, described the first prediction signal obtaining from 704 predictions of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel, and described mask value arrives according to described Depth Information Acquistion;

Coding unit 706 after prediction, prediction signal for each pixel described in the described current image block to be encoded of selecting to obtain according to described selected cell 706, coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.

In the embodiment of the present invention, acquiring unit 701, for obtaining current image block to be encoded, and obtains the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in described current image block to be encoded; First obtains unit 702, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of comparative result of the mask value of described two pixels that described acquiring unit 701 obtains; Second obtains unit 703, for according to the corresponding partitioned mode of comparative result of the mask value of described two pixels of described the first acquisition unit 702 acquisitions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded; Motion compensated prediction unit 704, be used for according to the motion vector information of described first subregion of described the second acquisition unit 703 acquisitions, described current image block to be encoded is carried out to motion compensated prediction, obtain described the first prediction signal, the motion vector information of described the second subregion obtaining according to described the second acquisition unit, described current image block to be encoded is carried out to motion compensated prediction, obtain described the second prediction signal; Selected cell 705, be used for according to the mask value of described current each pixel of image block to be encoded, described the first prediction signal obtaining from 704 predictions of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel; Coding unit 706 after prediction, prediction signal for each pixel described in the described current image block to be encoded of selecting to obtain according to described selected cell 706, coding after described current image block to be encoded is predicted, obtains the encoding code stream of described current image block to be encoded.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the code device that the embodiment of the present invention provides, in when coding, by the comparative result of mask value, determine that partitioned mode can determine partitioned mode fast, thereby improved the efficiency of the coding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 8, in another embodiment of the code device 70 that the embodiment of the present invention provides,

Described acquiring unit 701, also for obtaining at least one in the mask value of two other pixel of the pixel at described four angles except described two pixels;

Described first obtains unit 702, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of combination of the first comparative result and the second comparative result;

Described selected cell specifically for:

Consult Fig. 9, an embodiment of the decoding device 80 that the embodiment of the present invention provides comprises:

Acquiring unit 801, for from code stream, obtain current image block to be decoded prediction sign, described prediction sign is used to indicate the prediction that current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth, and according to depth information, obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be decoded;

First obtains unit 802, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of comparative result of the mask value of described two pixels that described acquiring unit 801 obtains;

Second obtains unit 803, for according to the corresponding partitioned mode of comparative result of the mask value of described two pixels of described the first acquisition unit 802 acquisitions, obtain the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded;

Motion compensated prediction unit 804, be used for according to the motion vector information of described first subregion of described the second acquisition unit 803 acquisitions, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal, according to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal;

Selected cell 805, be used for according to the mask value of described current each pixel of image block to be decoded, described the first prediction signal obtaining from 804 predictions of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel;

Decoding unit 806 after prediction, the prediction signal for each pixel described in the described current image block to be decoded of selecting to obtain according to described selected cell 805, obtains decoded image block.

During the embodiment of the present invention provides, acquiring unit 801 obtains the prediction sign of described code stream from the code stream of current image block to be decoded, and obtain the mask value of two pixels on same horizontal line or same vertical curve in the pixel at four angles in described current image block to be encoded, described prediction sign is used to indicate described code stream and has adopted the prediction of the piece subregion DBBP based on the degree of depth when coding, and indicates described current image block to be decoded to have two subregions; First obtains unit 802 according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of comparative result of the mask value of described two pixels that described acquiring unit 801 obtains; Second obtains unit 803 according to the corresponding partitioned mode of comparative result of the mask value of described two pixels of described the first acquisition unit 802 acquisitions, obtains the motion vector information of the first subregion and the motion vector information of the second subregion of described current image block to be encoded; The motion vector information of described the first subregion that motion compensated prediction unit 804 obtains according to described the second acquisition unit 803, described current image block to be decoded is carried out to motion compensated prediction, obtain described the first prediction signal, according to the motion vector information of described the second subregion, described current image block to be decoded is carried out to motion compensated prediction, obtain described the second prediction signal; Selected cell 805 is according to the mask value of each pixel in described current image block to be decoded, described the first prediction signal obtaining from 804 predictions of described motion compensated prediction unit and described the second prediction signal, select a prediction signal as the prediction signal of described each pixel, described mask value is from described the first prediction signal and described the second prediction signal, to select the sign as the prediction signal of described pixel; Described in the described current image block to be decoded that after prediction, decoding unit 806 is selected to obtain according to described selected cell 805, the prediction signal of each pixel, obtains decoded image block.With in prior art, by repeatedly carrying out addition, division and the operation such as relatively, could determine that partitioned mode compares, the decoding device that the embodiment of the present invention provides, in when decoding, by the comparative result of mask value, determine that partitioned mode can determine partitioned mode fast, thereby improved the efficiency of the decoding based on DBBP image block.

Alternatively, on the basis of embodiment corresponding to above-mentioned Fig. 9, in another embodiment of the decoding device 80 that the embodiment of the present invention provides,

Described acquiring unit 801, also for obtaining at least one in the mask value of two other pixel of the pixel at described four angles except described two pixels;

Described first obtains unit 802, for according to preset comparative result and the corresponding relation of partitioned mode, obtains the corresponding partitioned mode of combination of the first comparative result and the second comparative result;

Described selected cell 805 specifically for:

Figure 10 is the structural representation of embodiment of the present invention code device 50.Code device 50 can comprise input equipment 510, output equipment 520, processor 530 and memory 540.

Memory 540 can comprise read-only memory and random access memory, and provides instruction and data to processor 530.A part for memory 540 can also comprise nonvolatile RAM (NVRAM).

Memory 540 has been stored following element, executable module or data structure, or their subset, or their superset:

Operational order: comprise various operational orders, for realizing various operations.

Operating system: comprise various system programs, for realizing various basic businesses and processing hardware based task.

In embodiments of the present invention, processor 530, by calling the operational order (this operational order can be stored in operating system) of memory 540 storages, is carried out operation as follows:

Obtain current image block to be encoded;

In the embodiment of the present invention, code device 50 directly uses preset partitioned mode when coding, can determine fast partitioned mode, thereby improve the efficiency of the coding based on DBBP image block.

The operation of processor 530 control coding devices 50, processor 530 can also be called CPU (Central Processing Unit, CPU).Memory 540 can comprise read-only memory and random access memory, and provides instruction and data to processor 530.A part for memory 540 can also comprise nonvolatile RAM (NVRAM).In concrete application, each assembly of code device 50 is coupled by bus system 550, and wherein bus system 550, except comprising data/address bus, can also comprise power bus, control bus and status signal bus in addition etc.But for the purpose of clearly demonstrating, in the drawings various buses are all designated as to bus system 550.

The method that the invention described above embodiment discloses can be applied in processor 530, or is realized by processor 530.Processor 530 may be a kind of integrated circuit (IC) chip, has the disposal ability of signal.In implementation procedure, each step of said method can complete by the integrated logic circuit of the hardware in processor 530 or the instruction of form of software.Above-mentioned processor 530 can be general processor, digital signal processor (DSP), application-specific integrated circuit (ASIC) (ASIC), ready-made programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic device, discrete hardware components.Can realize or carry out disclosed each method, step and logic diagram in the embodiment of the present invention.General processor can be that microprocessor or this processor can be also the processors of any routine etc.Step in conjunction with the disclosed method of the embodiment of the present invention can directly be presented as that hardware decoding processor is complete, or complete with the hardware in decoding processor and software module combination.Software module can be positioned at random asccess memory, and flash memory, read-only memory, in the storage medium of this area maturations such as programmable read only memory or electrically erasable programmable memory, register.This storage medium is positioned at memory 540, and the information in processor 530 read memories 540 completes the step of said method in conjunction with its hardware.

Alternatively, described mask value is 0 or 1, processor 530 specifically can: when the mask value of described each pixel is 0, select described the first prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the second prediction signal as the prediction signal of described each pixel; Or,

Figure 11 is the structural representation of embodiment of the present invention decoding device 60.Decoding device 60 can comprise input equipment 610, output equipment 620, processor 630 and memory 640.

Memory 640 can comprise read-only memory and random access memory, and provides instruction and data to processor 630.A part for memory 640 can also comprise nonvolatile RAM (NVRAM).

Memory 640 has been stored following element, executable module or data structure, or their subset, or their superset:

In embodiments of the present invention, processor 630, by calling the operational order (this operational order can be stored in operating system) of memory 640 storages, is carried out operation as follows:

According to the prediction signal of each pixel described in described current image block to be decoded, the decoding after the code stream of described current image block to be decoded is predicted, obtains decoded image block.

In the embodiment of the present invention, decoding device 60 is directly used preset partitioned mode when decoding, can determine fast partitioned mode, thereby improve the efficiency of the decoding based on DBBP image block.

Processor 630 is controlled the operation of decoding device 60, and processor 630 can also be called CPU (Central Processing Unit, CPU).Memory 640 can comprise read-only memory and random access memory, and provides instruction and data to processor 630.A part for memory 640 can also comprise nonvolatile RAM (NVRAM).In concrete application, each assembly of decoding device 60 is coupled by bus system 650, and wherein bus system 650, except comprising data/address bus, can also comprise power bus, control bus and status signal bus in addition etc.But for the purpose of clearly demonstrating, in the drawings various buses are all designated as to bus system 650.

The method that the invention described above embodiment discloses can be applied in processor 630, or is realized by processor 630.Processor 630 may be a kind of integrated circuit (IC) chip, has the disposal ability of signal.In implementation procedure, each step of said method can complete by the integrated logic circuit of the hardware in processor 630 or the instruction of form of software.Above-mentioned processor 630 can be general processor, digital signal processor (DSP), application-specific integrated circuit (ASIC) (ASIC), ready-made programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic device, discrete hardware components.Can realize or carry out disclosed each method, step and logic diagram in the embodiment of the present invention.General processor can be that microprocessor or this processor can be also the processors of any routine etc.Step in conjunction with the disclosed method of the embodiment of the present invention can directly be presented as that hardware decoding processor is complete, or complete with the hardware in decoding processor and software module combination.Software module can be positioned at random asccess memory, and flash memory, read-only memory, in the storage medium of this area maturations such as programmable read only memory or electrically erasable programmable memory, register.This storage medium is positioned at memory 640, and the information in processor 630 read memories 640 completes the step of said method in conjunction with its hardware.

The described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.Alternatively, described mask value is 0 or 1, processor 630 specifically can: when the mask value of described each pixel is 0, select described the first prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the second prediction signal as the prediction signal of described each pixel; Or,

Figure 12 is the structural representation of embodiment of the present invention code device 70.Code device 70 can comprise input equipment 710, output equipment 720, processor 730 and memory 740.

Memory 740 can comprise read-only memory and random access memory, and provides instruction and data to processor 730.A part for memory 740 can also comprise nonvolatile RAM (NVRAM).

Memory 740 has been stored following element, executable module or data structure, or their subset, or their superset:

In embodiments of the present invention, processor 730, by calling the operational order (this operational order can be stored in operating system) of memory 740 storages, is carried out operation as follows:

In the embodiment of the present invention, code device 70 determines that by the comparative result of mask value partitioned mode can determine partitioned mode fast in when coding during at coding, thereby has improved the efficiency of the coding based on DBBP image block.

The operation of processor 730 control coding devices 70, processor 730 can also be called CPU (Central Processing Unit, CPU).Memory 740 can comprise read-only memory and random access memory, and provides instruction and data to processor 730.A part for memory 740 can also comprise nonvolatile RAM (NVRAM).In concrete application, each assembly of code device 70 is coupled by bus system 750, and wherein bus system 750, except comprising data/address bus, can also comprise power bus, control bus and status signal bus in addition etc.But for the purpose of clearly demonstrating, in the drawings various buses are all designated as to bus system 750.

The method that the invention described above embodiment discloses can be applied in processor 730, or is realized by processor 730.Processor 730 may be a kind of integrated circuit (IC) chip, has the disposal ability of signal.In implementation procedure, each step of said method can complete by the integrated logic circuit of the hardware in processor 730 or the instruction of form of software.Above-mentioned processor 730 can be general processor, digital signal processor (DSP), application-specific integrated circuit (ASIC) (ASIC), ready-made programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic device, discrete hardware components.Can realize or carry out disclosed each method, step and logic diagram in the embodiment of the present invention.General processor can be that microprocessor or this processor can be also the processors of any routine etc.Step in conjunction with the disclosed method of the embodiment of the present invention can directly be presented as that hardware decoding processor is complete, or complete with the hardware in decoding processor and software module combination.Software module can be positioned at random asccess memory, and flash memory, read-only memory, in the storage medium of this area maturations such as programmable read only memory or electrically erasable programmable memory, register.This storage medium is positioned at memory 740, and the information in processor 730 read memories 740 completes the step of said method in conjunction with its hardware.

Alternatively, described mask value is 0 or 1, processor 730 specifically can: when the mask value of described each pixel is 0, select described the first prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the second prediction signal as the prediction signal of described each pixel; Or,

Alternatively, described processor 730 also can obtain at least one in the mask value of two other pixel except described two pixels in the pixel at described four angles;

Described processor 730 specifically can, according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of combination of the first comparative result and the second comparative result;

Figure 13 is the structural representation of embodiment of the present invention decoding device 80.Decoding device 80 can comprise input equipment 810, output equipment 820, processor 830 and memory 840.

Memory 840 can comprise read-only memory and random access memory, and provides instruction and data to processor 830.A part for memory 840 can also comprise nonvolatile RAM (NVRAM).

Memory 840 has been stored following element, executable module or data structure, or their subset, or their superset:

In embodiments of the present invention, processor 830, by calling the operational order (this operational order can be stored in operating system) of memory 840 storages, is carried out operation as follows:

From code stream, obtain current image block to be decoded prediction sign, described prediction sign is used to indicate the prediction that current image block to be decoded has adopted the piece subregion DBBP based on the degree of depth

In the embodiment of the present invention, decoding device 80 determines that by the comparative result of mask value partitioned mode can determine partitioned mode fast in when decoding, thereby has improved the efficiency of the coding based on DBBP image block.

Processor 830 is controlled the operation of decoding device 80, and processor 830 can also be called CPU (Central Processing Unit, CPU).Memory 840 can comprise read-only memory and random access memory, and provides instruction and data to processor 830.A part for memory 840 can also comprise nonvolatile RAM (NVRAM).In concrete application, each assembly of decoding device 80 is coupled by bus system 850, and wherein bus system 850, except comprising data/address bus, can also comprise power bus, control bus and status signal bus in addition etc.But for the purpose of clearly demonstrating, in the drawings various buses are all designated as to bus system 850.

The method that the invention described above embodiment discloses can be applied in processor 830, or is realized by processor 830.Processor 830 may be a kind of integrated circuit (IC) chip, has the disposal ability of signal.In implementation procedure, each step of said method can complete by the integrated logic circuit of the hardware in processor 830 or the instruction of form of software.Above-mentioned processor 830 can be general processor, digital signal processor (DSP), application-specific integrated circuit (ASIC) (ASIC), ready-made programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic device, discrete hardware components.Can realize or carry out disclosed each method, step and logic diagram in the embodiment of the present invention.General processor can be that microprocessor or this processor can be also the processors of any routine etc.Step in conjunction with the disclosed method of the embodiment of the present invention can directly be presented as that hardware decoding processor is complete, or complete with the hardware in decoding processor and software module combination.Software module can be positioned at random asccess memory, and flash memory, read-only memory, in the storage medium of this area maturations such as programmable read only memory or electrically erasable programmable memory, register.This storage medium is positioned at memory 840, and the information in processor 830 read memories 840 completes the step of said method in conjunction with its hardware.

Alternatively, described mask value is 0 or 1, processor 830 specifically can: when the mask value of described each pixel is 0, select described the first prediction signal as the prediction signal of described each pixel, when the mask value of described each pixel is 1, select described the second prediction signal as the prediction signal of described each pixel; Or,

Alternatively, described processor 830 also can obtain at least one in the mask value of two other pixel except described two pixels in the pixel at described four angles;

Described processor 830 specifically can, according to preset comparative result and the corresponding relation of partitioned mode, obtain the corresponding partitioned mode of combination of the first comparative result and the second comparative result;

One of ordinary skill in the art will appreciate that all or part of step in the whole bag of tricks of above-described embodiment is to come the hardware that instruction is relevant (for example processor) to complete by program, this program can be stored in a computer-readable recording medium, and storage medium can comprise: ROM, RAM, disk or CD etc.

The coding above embodiment of the present invention being provided, coding/decoding method and device are described in detail, applied specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment is just for helping to understand method of the present invention and core concept thereof; , for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention meanwhile.

Claims

1. a coding method for the image block based on depth information, is characterized in that, comprising:

Obtain current image block to be encoded;

2. coding method according to claim 1, is characterized in that, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

3. coding method according to claim 1 and 2, is characterized in that, described mask value is 0 or 1,

4. a coding/decoding method for the image block based on depth information, is characterized in that, comprising:

5. coding/decoding method according to claim 4, is characterized in that, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

6. according to the coding/decoding method described in claim 4 or 5, it is characterized in that, described mask value is 0 or 1,

7. a coding method for the image block based on depth information, is characterized in that, comprising:

8. coding method according to claim 7, it is characterized in that, during the described mask value according to two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion, described method also comprises:

9. according to the coding method described in claim 7 or 8, it is characterized in that, the described partitioned mode in the corresponding relation of described preset comparative result and partitioned mode comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

10. according to the arbitrary described coding method of claim 7-9, it is characterized in that, described mask value is 0 or 1,

The coding/decoding method of 11. 1 kinds of image blocks based on depth information, is characterized in that, comprising:

12. coding/decoding methods according to claim 11, it is characterized in that, during the described mask value according to two pixels on same horizontal line or same vertical curve in the pixel at four angles in current image block to be encoded described in Depth Information Acquistion, described method also comprises:

13. according to the coding method described in claim 11 or 12, it is characterized in that, the described partitioned mode in the corresponding relation of described preset comparative result and partitioned mode comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

14. according to the arbitrary described coding method of claim 11-13, it is characterized in that, described mask value is 0 or 1,

15. 1 kinds of code devices, is characterized in that, comprising:

Acquiring unit, for obtaining current image block to be encoded;

16. code devices according to claim 15, is characterized in that, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

17. according to the code device described in claim 15 or 16, it is characterized in that, described mask value is 0 or 1,

Described selected cell specifically for:

18. 1 kinds of decoding devices, is characterized in that, comprising:

19. decoding devices according to claim 18, is characterized in that, the described preset partitioned mode that comprises two subregions comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

20. according to the decoding device described in claim 18 or 19, it is characterized in that, described mask value is 0 or 1,

Described selected cell specifically for:

21. 1 kinds of code devices, is characterized in that, comprising:

22. code devices according to claim 21, is characterized in that,

23. according to the code device described in claim 21 or 22, it is characterized in that, described preset comparative result and partitioned mode comprise two subregions that are arranged above and below, or two subregions of left and right arrangement.

24. according to the arbitrary described code device of claim 21-23, it is characterized in that, described mask value is 0 or 1,

Described selected cell specifically for:

25. 1 kinds of decoding devices, is characterized in that, comprising:

26. decoding devices according to claim 25, is characterized in that,

27. according to the decoding device described in claim 25 or 26, it is characterized in that, the described partitioned mode in the corresponding relation of described preset comparative result and partitioned mode comprises two subregions that are arranged above and below, or two subregions of left and right arrangement.

28. according to the arbitrary described decoding device of claim 25-27, it is characterized in that, described mask value is 0 or 1,

Described selected cell specifically for: