CN104104956B

CN104104956B - For layered video coding and the method for decoding, encoding apparatus and decoding apparatus

Info

Publication number: CN104104956B
Application number: CN201310119503.6A
Authority: CN
Inventors: 魏湘辉; 昝劲文
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2013-04-08
Filing date: 2013-04-08
Publication date: 2017-10-17
Anticipated expiration: 2033-04-08
Also published as: CN104104956A

Abstract

The embodiments of the invention provide a kind of for layered video coding and the method for decoding, encoding apparatus and decoding apparatus.This method includes：The associated picture block of enhancement layer image block is determined, the associated picture block includes the corresponding Primary layer image block of enhancement layer image block or the encoded image block adjacent with the enhancement layer image block；Obtain the corresponding residual information of associated picture block；According to the corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, whether vague generalization residual prediction algorithm is used, wherein, the corresponding residual information of associated picture block is used to indicate whether the associated picture block has residual error or the residual error size degree of the associated picture block.The method for being used for layered video coding and decoding of the embodiment of the present invention, encoding apparatus and decoding apparatus can reduce the complexity of coding and decoding video and reduce the time of encoding and decoding.

Description

For layered video coding and the method for decoding, encoding apparatus and decoding apparatus

Technical field

The present invention relates to the communications field, and more particularly, to it is a kind of for layered video coding and decoding method, Encoding apparatus and decoding apparatus.

Background technology

In recent years, with the popularization and development of internet and the communication technology, digital media technology by it is extensive must be applied to it is many Individual field.Meanwhile, the appearance of high definition and ultra high-definition video proposes higher requirement to video compression technology.High-performance is regarded Frequency coding (High Efficiency Video Coding, HEVC) video standard adapts to this demand and is formulated.

Scalable high-performance video coding (Scalable High-efficiency Video Coding, SHVC) is Extension of the HEVC video encoding standards on scalable video in a formulation.In SHVC, video code flow contains multilayer Video data.The video data (Layer ID=0) of the bottom is referred to as on Primary layer (Base Layer, BL), Primary layer Data hierarchy (Layer ID>0) it is referred to as enhancement layer (Enhancement Layer, EL).So, in different application scene In, the vision signal of different resolution and different images quality can be provided by extracting the data adaptive of different levels.

In SHVC, when carrying out inter prediction or motion compensation, vague generalization residual prediction (Generalized can be used Residual Prediction, GRP) algorithm.GRP algorithms use base layer data amendment enhancement layer prediction signal, contribute to Maximum code rate gains in SHVC.However, GRP also considerably increases the complexity of encoding and decoding, the time of encoding and decoding is extended.

The content of the invention

The embodiments of the invention provide a kind of method for layered video coding and the method for decoding, code device is conciliate Code device, can reduce the complexity of coding and decoding video and reduce the time of encoding and decoding.

First aspect there is provided a kind of method for layered video coding, including：Determine the correlation of enhancement layer image block Image block, the associated picture block includes the corresponding Primary layer image block of enhancement layer image block or adjacent with the enhancement layer image block Encoded image block；Obtain the corresponding residual information of associated picture block；According to the corresponding residual information of associated picture block, really When being scheduled on to enhancement layer image block execution inter prediction, if use vague generalization residual prediction algorithm, in order to increase to described Strong tomographic image block performs the inter prediction to obtain the prediction signal of the enhancement layer image block, and according to the enhancement layer figure As the prediction signal of block is encoded to the enhancement layer image block, wherein, the corresponding residual information of associated picture block is used for Indicate whether the associated picture block has residual error or the residual error size degree of the associated picture block.

With reference in a first aspect, in the first possible implementation of first aspect, the associated picture block is corresponding residual Poor information includes the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block At least one of residual error non-zero transform coefficient included with one or more transform blocks of the associated picture block, wherein, it is described The coded block flag of associated picture block is used to indicate whether the associated picture block has a residual error, one of the associated picture block or The coded block flag of multiple transform blocks is used to indicate whether corresponding transform block has residual error.

With reference to the first possible implementation of first aspect, in second of possible implementation of first aspect In, this is according to the corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, if Using vague generalization residual prediction algorithm, including：

When the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, and the correlation is schemed When indicating that the associated picture block has residual error as the coded block flag of block, it is determined that performing inter prediction to the enhancement layer image block When, using the vague generalization residual prediction algorithm；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates that corresponding transform block includes residual error During non-zero transform coefficient, it is determined that when performing inter prediction to the enhancement layer image block, using the vague generalization residual prediction algorithm； Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block is big It is more than the second threshold in first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of value is met, it is determined that when performing inter prediction to the enhancement layer image block, it is residual using the vague generalization Poor prediction algorithm.

With reference to the first possible implementation of first aspect, in the third possible implementation of first aspect In, this is according to the corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, if Using vague generalization residual prediction algorithm, including：

When the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, and the correlation is schemed When indicating the associated picture block without residual error as the coded block flag of block, it is determined that performing inter prediction to the enhancement layer image block When, the vague generalization residual prediction algorithm is not used；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block not comprising residual During poor non-zero transform coefficient, it is determined that when performing inter prediction to the enhancement layer image block, the vague generalization residual prediction is not used Algorithm；Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block is small It is less than the second threshold in first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of value is met, it is determined that when performing inter prediction to the enhancement layer image block, the vague generalization is not used Residual prediction algorithm.

With reference to the third possible implementation of first aspect, in the 4th kind of possible implementation of first aspect In, the associated picture block includes the two or more encoded image blocks adjacent with the enhancement layer image block.

With reference in a first aspect, the first possible implementation of first aspect, second of possible reality of first aspect Existing mode, the 4th kind of possible implementation of the third possible implementation or first aspect of first aspect, first In 5th kind of possible implementation of aspect, this method also includes：, will it is determined that during using the vague generalization residual prediction algorithm In vague generalization residual prediction flag position write-in enhancement layer bitstream, wherein, the vague generalization write in the enhancement layer bitstream is residual Poor prediction indication position is set to indicate that the pattern of the vague generalization residual prediction algorithm；It is determined that not using the vague generalization residual error pre- During method of determining and calculating, not by vague generalization residual prediction flag position write-in enhancement layer bitstream.

Second aspect there is provided it is a kind of for layered video decode method, including：

The associated picture block of enhancement layer image block is determined, the associated picture block includes the corresponding Primary layer of enhancement layer image block Image block or the decoded image blocks adjacent with the enhancement layer image block；

Obtain the corresponding residual information of the enhancement layer image block；

According to the corresponding residual information of associated picture block, it is determined that to the enhancement layer image block perform decoding reconstruction operation In motion compensation when, if vague generalization residual prediction algorithm is used, in order to perform the fortune to the enhancement layer image block Dynamic compensation is to obtain the prediction signal of the enhancement layer image block, and according to the prediction signal of the enhancement layer image block to described Enhancement layer image block is decoded, wherein, the corresponding residual information of associated picture block is used to indicate the associated picture block in solution Whether there is the residual error size degree of residual error or the associated picture block after the decoding after code.

With reference to second aspect, in the first possible implementation of second aspect, the associated picture block is corresponding residual Poor information includes the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block At least one of residual error non-zero transform coefficient included with one or more transform blocks of the associated picture block.

With reference to the first possible implementation of second aspect, in second of possible implementation of second aspect In, this is according to the corresponding residual information of associated picture block, it is determined that in the enhancement layer image block perform decoding reconstruction operation Motion compensation when, if use vague generalization residual prediction algorithm, including：

When the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, and the correlation is schemed When indicating that the associated picture block has residual error after the decoding as the coded block flag of block, it is determined that performing solution to the enhancement layer image block Code reconstruction operation in motion compensation when, using the vague generalization residual prediction algorithm；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates that corresponding transform block includes residual error During non-zero transform coefficient, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, using this one As change residual prediction algorithm；Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block is big It is more than the second threshold in first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of value is met, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, using the vague generalization residual prediction algorithm.

With reference to the first possible implementation of second aspect, in the third possible implementation of second aspect In, this is according to the corresponding residual information of associated picture block, it is determined that in the enhancement layer image block perform decoding reconstruction operation Motion compensation when, if use vague generalization residual prediction algorithm, including：

When the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, and the correlation is schemed When indicating the associated picture block after the decoding without residual error as the coded block flag of block, it is determined that performing solution to the enhancement layer image block Code reconstruction operation in motion compensation when, the vague generalization residual prediction algorithm is not used；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block not comprising residual During poor non-zero transform coefficient, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, not using The vague generalization residual prediction algorithm；Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block is small It is less than the second threshold in first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of value is met, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, the vague generalization residual prediction algorithm is not used；

With reference to the third possible implementation of second aspect, in the 4th kind of possible implementation of second aspect In, the associated picture block includes the two neighboring or more than two decoded image blocks with the enhancement layer image block.

With reference to second aspect, the first possible implementation of second aspect, second of possible reality of second aspect Existing mode, the 4th kind of possible implementation of the third possible implementation or second aspect of second aspect, second In 5th kind of possible implementation of aspect, this method also includes：It is determined that during using the vague generalization residual prediction algorithm, from Vague generalization residual prediction flag position is parsed in enhancement layer bitstream, to be determined according to the setting of vague generalization residual prediction flag position The pattern of the vague generalization residual prediction algorithm；When it is determined that not using the vague generalization residual prediction algorithm, not from enhancement layer code Vague generalization residual prediction flag position is parsed in stream.

The third aspect there is provided a kind of code device, including：

First determining unit, the associated picture block for determining enhancement layer image block, associated picture block includes enhancement layer The corresponding Primary layer image block of image block or the encoded image block adjacent with the enhancement layer image block；

Acquiring unit, for obtaining the corresponding residual information of the associated picture block；

Second determining unit, for according to the corresponding residual information of associated picture block, it is determined that to the enhancement layer image It is described in order to be performed to the enhancement layer image block when block performs inter prediction, if use vague generalization residual prediction algorithm Inter prediction is to obtain the prediction signal of the enhancement layer image block, and according to the prediction signal of the enhancement layer image block to institute Enhancement layer image block is stated to be encoded, wherein, the corresponding residual information of associated picture block is used to indicate that the associated picture block is It is no to have residual error or the residual error size degree of the associated picture block.

With reference to the third aspect, in the first possible implementation of the third aspect, the associated picture block is corresponding residual Poor information includes the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block At least one of residual error non-zero transform coefficient included with one or more transform blocks of the associated picture block, wherein, it is described The coded block flag of associated picture block is used to indicate whether the associated picture block has a residual error, one of the associated picture block or The coded block flag of multiple transform blocks is used to indicate whether corresponding transform block has residual error.

With reference to the first possible implementation of the third aspect, in second of possible implementation of the third aspect In, second determining unit specifically for：

With reference to the first possible implementation of the third aspect, in the third possible implementation of the third aspect In, second determining unit specifically for：

With reference to the third possible implementation of the third aspect, in the 4th kind of possible implementation of the third aspect In, the associated picture block includes the two or more encoded image blocks adjacent with the enhancement layer image block.

With reference to the third aspect, the first possible implementation of the third aspect, second of possible reality of the third aspect Existing mode, the 4th kind of possible implementation of the third possible implementation or the third aspect of the third aspect, the 3rd In 5th kind of possible implementation of aspect, the code device also includes writing unit, for true in second determining unit When using the vague generalization residual prediction algorithm surely, vague generalization residual prediction flag position is write in enhancement layer bitstream, wherein, it is described The vague generalization residual prediction flag position write in enhancement layer bitstream is set to indicate that the vague generalization residual prediction algorithm Pattern, when second determining unit determines not using the vague generalization residual prediction algorithm, not by vague generalization residual prediction flag In position write-in enhancement layer bitstream.

Fourth aspect there is provided a kind of decoding apparatus, including：

First determining unit, the associated picture block for determining enhancement layer image block, associated picture block includes enhancement layer The corresponding Primary layer image block of image block or the decoded image blocks adjacent with the enhancement layer image block；

Acquiring unit, for obtaining the corresponding residual information of the enhancement layer image block；

Second determining unit, for according to the corresponding residual information of associated picture block, it is determined that to the enhancement layer image During motion compensation in block perform decoding reconstruction operation, if vague generalization residual prediction algorithm is used, in order to the enhancing Tomographic image block performs the motion compensation to obtain the prediction signal of the enhancement layer image block, and according to the enhancement layer image The prediction signal of block is decoded to the enhancement layer image block, wherein, the corresponding residual information of associated picture block is used to refer to Show whether the associated picture block has the residual error size degree of residual error or the associated picture block after the decoding after the decoding.

With reference to fourth aspect, in the first possible implementation of fourth aspect, the associated picture block is corresponding residual Poor information includes the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block At least one of residual error non-zero transform coefficient included with one or more transform blocks of the associated picture block.

With reference to the first possible implementation of fourth aspect, in second of possible implementation of fourth aspect In, second determining unit specifically for：

With reference to the first possible implementation of fourth aspect, in the third possible implementation of fourth aspect In, second determining unit specifically for：

With reference to the third possible implementation of fourth aspect, in the 4th kind of possible implementation of fourth aspect In, the associated picture block includes the two neighboring or more than two decoded image blocks with the enhancement layer image block.

With reference to fourth aspect, the first possible implementation of fourth aspect, second of possible reality of fourth aspect Existing mode, the 4th kind of possible implementation of the third possible implementation or fourth aspect of fourth aspect, the 4th In 5th kind of possible implementation of aspect, the decoding apparatus also includes resolution unit, for true in second determining unit When using the vague generalization residual prediction algorithm surely, vague generalization residual prediction flag position is parsed from enhancement layer bitstream, with according to institute The setting for stating vague generalization residual prediction flag position determines the pattern of the vague generalization residual prediction algorithm, in second determining unit It is determined that when not using the vague generalization residual prediction algorithm, vague generalization residual prediction flag position is parsed not from enhancement layer bitstream.

There is provided a kind of code device, including memory and processor for 5th aspect；Store program codes in memory, Processor calls the program code stored in memory to perform following operate：

The associated picture block of enhancement layer image block is determined, the associated picture block includes the corresponding Primary layer of enhancement layer image block Image block or the encoded image block adjacent with the enhancement layer image block；

Obtain the corresponding residual information of the associated picture block；

According to the corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, Whether vague generalization residual prediction algorithm is used, it is described to obtain in order to perform the inter prediction to the enhancement layer image block The prediction signal of enhancement layer image block, and the enhancement layer image block is carried out according to the prediction signal of the enhancement layer image block Coding, wherein, the corresponding residual information of associated picture block is used to indicate whether the associated picture block has residual error or correlation figure As the residual error size degree of block.

With reference to the 5th aspect, in the first possible implementation of the 5th aspect, the associated picture block is corresponding residual Poor information includes the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block At least one of residual error non-zero transform coefficient included with one or more transform blocks of the associated picture block, wherein, it is described The coded block flag of associated picture block is used to indicate whether the associated picture block has a residual error, one of the associated picture block or The coded block flag of multiple transform blocks is used to indicate whether corresponding transform block has residual error.

With reference to the first possible implementation of the 5th aspect, in second of possible implementation of the 5th aspect In, processor calls the program code stored in memory specifically to perform following operate：

With reference to the first possible implementation of the 5th aspect, in the third possible implementation of the 5th aspect In, this calls the program code stored in memory specifically to perform according to the corresponding residual information of associated picture block, processor Operate below：

With reference to the third possible implementation of the 5th aspect, in the 4th kind of possible implementation of the 5th aspect In, the associated picture block includes the two or more encoded image blocks adjacent with the enhancement layer image block.

With reference to the 5th aspect, the first possible implementation of the 5th aspect, second of possible reality of the 5th aspect 4th kind of possible implementation of existing mode, the third possible implementation of the 5th aspect or the 5th aspect, the 5th In 5th kind of possible implementation of aspect, processor calls the program code stored in memory also to perform following operation：

It is determined that during using the vague generalization residual prediction algorithm, by vague generalization residual prediction flag position write-in enhancement layer bitstream In, wherein, the vague generalization residual prediction flag position write in the enhancement layer bitstream is set to indicate that the vague generalization is residual The pattern of poor prediction algorithm；When it is determined that not using the vague generalization residual prediction algorithm, not by vague generalization residual prediction flag position Write in enhancement layer bitstream.

There is provided a kind of decoding apparatus, including memory and processor for 6th aspect；Store program codes in memory, Processor calls the program code stored in memory to perform following operate：

With reference to the 6th aspect, in the first possible implementation of the 6th aspect, the associated picture block is corresponding residual Poor information includes the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block At least one of residual error non-zero transform coefficient included with one or more transform blocks of the associated picture block.

With reference to the first possible implementation of the 6th aspect, in second of possible implementation of the 6th aspect In, processor calls the program code stored in memory specifically to perform following operate：

With reference to the first possible implementation of the 6th aspect, in the third possible implementation of the 6th aspect In, this calls the program code stored in memory specifically to perform according to the corresponding residual information of associated picture block, processor Operate below：

With reference to the third possible implementation of the 6th aspect, in the 4th kind of possible implementation of the 6th aspect In, the associated picture block includes the two neighboring or more than two decoded image blocks with the enhancement layer image block.

With reference to the 6th aspect, the first possible implementation of the 6th aspect, second of possible reality of the 6th aspect 4th kind of possible implementation of existing mode, the third possible implementation of the 6th aspect or the 6th aspect, the 6th In 5th kind of possible implementation of aspect, processor calls the program code stored in memory also to perform following operation：

It is determined that during using the vague generalization residual prediction algorithm, vague generalization residual prediction flag is parsed from enhancement layer bitstream Position, to determine the pattern of the vague generalization residual prediction algorithm according to the setting of vague generalization residual prediction flag position；True When determining not using the vague generalization residual prediction algorithm, vague generalization residual prediction flag position is parsed not from enhancement layer bitstream.

Therefore, in embodiments of the present invention, only in the residual information according to associated picture block, it is determined that needing to use GRP During algorithm, just using GRP algorithm performs inter prediction or motion compensation, so as to ensure picture quality substantially not by In the case of influence, reduce coding and decoding video complexity and reduce the encoding and decoding time.

Brief description of the drawings

In order to illustrate the technical solution of the embodiments of the present invention more clearly, will make below to required in the embodiment of the present invention Accompanying drawing is briefly described, it should be apparent that, drawings described below is only some embodiments of the present invention, for For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings Accompanying drawing.

Fig. 1 is the schematic figure of GRP algorithms.

Fig. 2 is the indicative flowchart of the method for layered video coding according to embodiments of the present invention.

Fig. 3 is the schematic figure of the enhancement layer image block according to another embodiment of the present invention image block adjacent with its.

Fig. 4 is the indicative flowchart of the method according to another embodiment of the present invention decoded for layered video.

Fig. 5 is the schematic block diagram of code device according to another embodiment of the present invention.

Fig. 6 is the schematic block diagram of code device according to another embodiment of the present invention.

Fig. 7 is the schematic block diagram of decoding apparatus according to another embodiment of the present invention.

Fig. 8 is the schematic block diagram of decoding apparatus according to another embodiment of the present invention.

Fig. 9 is the schematic block diagram of code device according to another embodiment of the present invention.

Figure 10 is the schematic block diagram of decoding apparatus according to another embodiment of the present invention.

Figure 11 is the schematic block diagram of decoding apparatus according to another embodiment of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation is described, it is clear that described embodiment is a part of embodiment of the present invention, rather than whole embodiments.Based on this hair Embodiment in bright, the every other reality that those of ordinary skill in the art are obtained on the premise of creative work is not made Example is applied, should all belong to the scope of protection of the invention.

In order to be more clearly understood from the present invention, the GRP algorithms mentioned in the embodiment of the present invention are entered below with reference to Fig. 1 Row is described in detail.

Fig. 1 shows the schematic diagram of GRP algorithms.Wherein, CurPic is current frame image, and RefPic is reference frame image； Be is that enhancement layer is performed inter prediction or the image block of motion compensation, and Pe is enhancement layer prediction signal, and MVe is corresponding with Pe Enhancement layer motion vector；Equally, Bb is the image block that Primary layer corresponds to Be, and Pb is Primary layer prediction signal, Primary layer Motion vector is using the value as enhancement layer, i.e. MVe.

GRP algorithms are divided into three steps：

1) image block (Be) that enhancement layer is performed inter prediction or motion compensation is searched in enhancement layer reference frame Refpic Rope obtains prediction signal Pe to the image block most matched, while obtaining motion vector MVe；

2) Primary layer correspondence image block Bb uses enhancement layer Be motion vector MVe, obtains pre- in Primary layer reference frame Survey signal Pb.

3) Pe is according to formula P=P_e+w(B_b-P_b) be modified, final prediction signal P is obtained, wherein w can be 0,0.5, 1, it is known respectively as Res-0 patterns, Res-0.5 patterns, Res-1 patterns.

For some special circumstances, using Primary layer correspondence image block Bb as prediction signal, better performance can be obtained, So as to according to formula P=B_b+0.5(P_e-P_b) calculate, this pattern is referred to as Diff-0.5 patterns.

In summary, the pattern that GRP algorithms are modified using base layer data to the prediction signal of enhancement layer, is divided into Pattern in 4：Res-0, Res-0.5, Res-1 and Diff-0.5.Correspondingly, can be using GRP marks in the grammer of coding unit Will position (bit of length 2) indicates corresponding pattern, for example, GRP flag bits represent Res-0/Res-0.5/ respectively with 0/1/2/3 Res-1/Diff-0.5 patterns, wherein, coding unit (Coding Unit, CU) is that have phase inside the base unit of coding, CU Same prediction mode.CU is also referred to as image block in the present invention.Prediction mode includes inter prediction (Inter Prediction) With infra-frame prediction (Intra Prediction).One CU can be divided into multiple predicting units (Prediction Unit, PU), PU is the base unit of perform prediction.One CU can be divided into multiple converter units (Transform Unit, TU), TU It is the base unit for performing conversion.

The method for layered video coding to the embodiment of the present invention is described in detail below.Fig. 1 is according to this hair The indicative flowchart of the method 100 for layered video coding of bright embodiment.This method 100 can be held by code device OK.As shown in figure 1, this method 100 includes：

S110, determines the associated picture block of enhancement layer image block, and the associated picture block includes enhancement layer image block correspondence Primary layer image block or the encoded image block adjacent with the enhancement layer image block；

S120, obtains the corresponding residual information of associated picture block, according to the corresponding residual information of associated picture block, really When being scheduled on to enhancement layer image block execution inter prediction, if GRP algorithms are used, in order to be held to the enhancement layer image block The row inter prediction is believed with obtaining the prediction signal of the enhancement layer image block according to the prediction of the enhancement layer image block Number the enhancement layer image block is encoded, wherein, the corresponding residual information of associated picture block is used to indicate correlation figure As whether block has residual error or the residual error size degree of the associated picture block.

In SHVC, due in inter prediction, code rate gains maximum in SHVC are although contribute to using GRP algorithms, But the complexity of coding is also considerably increased, the time of coding is extended, it is therefore desirable to it is determined that being calculated under what circumstances using GRP Method and which kind of in the case of without using GRP algorithms, can ensure picture quality while, reduce coding complexity, reduction compile The time of code.

Specifically, code device is when needing to perform inter prediction to current EL image block, it may be determined that the enhancing The associated picture block of tomographic image block, wherein, the associated picture block can include the corresponding Primary layer of encoding enhancement layer image block Image block or the encoded image block adjacent with the enhancement layer image block；Then, code device can obtain the associated picture block Corresponding residual information, wherein, the corresponding residual information of associated picture block is used to indicate whether the associated picture block has residual error Or the residual error size degree of the associated picture block；After the residual information of the associated picture block is got, code device can root According to the residual information of the associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, if calculated using GRP Method；When it is determined that needing to use GRP algorithms, then when performing inter prediction to the enhancement layer image block, GRP algorithms are used to obtain The prediction signal of enhancement layer image block is got, can be pre- using the interframe of single-layer video when it is determined that GRP algorithms need not be used Survey pattern performs inter prediction to get the prediction signal of enhancement layer image block, i.e., when carrying out inter prediction, it is not necessary to adopt With base layer data amendment enhancement layer prediction signal., can be according to the increasing after the prediction signal of enhancement layer image block is got The prediction signal of strong tomographic image block, coding is performed to the enhancement layer image block, for example, the residual error of the enhancement layer image block is obtained, Then orthogonal transformation, quantization and entropy code etc. are performed.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag Include the corresponding Primary layer image block of the enhancement layer image block or the encoded image block adjacent with the enhancement layer image block, and according to The corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, if calculated using GRP Method, wherein, the corresponding residual information of associated picture block is used to indicate whether the associated picture block has residual error or be somebody's turn to do in encoded The residual error size degree of associated picture block, so that, only in the residual information according to associated picture block, it is determined that needing to use GRP During algorithm, inter prediction is just performed to enhancement layer image block using GRP algorithms, so as to not have substantially in guarantee picture quality In the case of being affected, reduce video encoding complexity and reduce video encoding time.

In embodiments of the present invention, image block is the elementary cell of encoding and decoding.Image block can be divided into one or more Transform block.And transform block be store non-zero residual transform coefficient elementary cell, including brightness (luma) transform block, blueness it is poor (chroma blue, cb) transform block and poor (chroma red, the cr) transform block of red.

In embodiments of the present invention, the encoded image block adjacent with enhancement layer image block can be with as shown in figure 3, include Upper left-hand image block, upper images block, upper right side image block, left image block and the lower left image of the enhancement layer image block Block.The associated picture block of the enhancement layer image block can include two or more (examples adjacent with the enhancement layer image block Such as, upper left-hand image block and upper images block) encoded image block.It is of course also possible to only include an encoded image block. The embodiment of the present invention is defined not to this.Certainly, the encoded image block adjacent with enhancement layer image block can also be it Its image block, for example, upper left-hand image block, upper images block, upper right side image block, right image block and lower right image block； Depending on specifically can be according to coded sequence.

In embodiments of the present invention, the corresponding residual information of associated picture block of enhancement layer image block can include：The phase Close image block coded block flag (coding block flag, rqt_root_cbf), the associated picture block it is one or more In the residual error non-zero transform coefficient that one or more transform blocks of the coded block flag of transform block and the associated picture block include At least one, wherein, the coded block flag of the associated picture block is used for whether having residual error after indicating associated picture block coding, should The coded block flag of one or more transform blocks of associated picture block is used for whether having residual error after indicating corresponding transform block coding. Wherein, the coded block flag of the image block can be residual error quaternary tree root coded block flag (the residual quad of the image block Tree root coding block flag, rqt_root_cbf), for example, rqt_root_cbf represents the associated picture for 0 Block represents that the associated picture block has residual error without residual error, rqt_root_cbf for 1.One or more transform blocks of the associated picture block Coded block flag refer to the associated picture block luminance transformation block coded block flag (cbf_luma), blueness difference transform block One or more of coded block flag (cbf_cb) and the coded block flag (cbf_cr) of red poor transform block.Cbf_luma is used In indicating whether the luminance transformation block includes residual error non-zero transform coefficient, for example, cbf_luma represents the luminance transformation block bag for 1 Non-zero transform coefficient containing residual error, cbf_luma is 0 expression luminance transformation block not comprising residual error non-zero transform coefficient.Cbf_cr is used In indicating whether red poor transform block includes residual error non-zero transform coefficient, for example, cbf_cr represents the poor transform block bag of the red for 1 Non-zero transform coefficient containing residual error, cbf_cr is that the 0 poor transform block of the expression red does not include residual error non-zero transform coefficient.Cbf_cb is used In indicating whether the poor transform block of the blueness includes residual error non-zero transform coefficient, for example, cbf_cb is the 1 poor transform block of the expression blueness Comprising residual error non-zero transform coefficient, cbf_cb is that the 0 poor transform block of the expression blueness does not include residual error non-zero transform coefficient.In this hair In bright embodiment, residual error non-zero transform coefficient can characterize the size degree of residual error, can specifically be characterized by number and amplitude The size degree of residual error.

In embodiments of the present invention, it is determined that during using GRP algorithms, code device, which can write GRP flag bits, to be strengthened In layer bit stream, for example, GRP flag bits represent Res-0/Res-0.5/Res-1/Diff-0.5 patterns respectively with 0/1/2/3, from And, decoding apparatus can determine to need the GRP algorithms using which kind of pattern according to the GRP flag bits；It is determined that not using GRP to calculate During method, GRP flag bits can not be write in enhancement layer bitstream, so as to save code stream.

In an embodiment of the present invention, according to the corresponding residual information of associated picture block described in S120, it is determined that When performing inter prediction to the enhancement layer image block, if use GRP algorithms, can include：In associated picture block correspondence Residual information including the associated picture block coded block flag when, and the coded block flag of the associated picture block indicates the correlation When image block has residual error, it is determined that when performing inter prediction to the enhancement layer image block, using the GRP algorithms；Or, in the phase It (can be one or many specified to close one or more transform blocks of the corresponding residual information of image block including the associated picture block Individual transform block or any one or more transform blocks) coded block flag when, and one or many of the associated picture block When the coded block flag of individual transform block indicates that corresponding transform block includes residual error non-zero transform coefficient, it is determined that to the enhancement layer figure When performing inter prediction as block, using the GRP algorithms；Or, include the correlation in the corresponding residual information of associated picture block One or more transform blocks of image block (can be one or more transform blocks for specifying or any one or more Transform block) include residual error non-zero transform coefficient when, and one or more transform blocks of the associated picture block residual error non-zero become The amplitude for changing coefficient is more than first threshold and the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When number is met more than at least one of Second Threshold, it is determined that when performing inter prediction to the enhancement layer image block, Using the GRP algorithms.If being unsatisfactory for above-mentioned three kinds of conditions, when carrying out inter prediction, GRP algorithms can not be used.

In another embodiment of the invention, described in S120 according to the corresponding residual information of associated picture block, really When being scheduled on to enhancement layer image block execution inter prediction, if use GRP algorithms, can include：In the associated picture block pair When the residual information answered includes the coded block flag of the associated picture block, and the coded block flag of the associated picture block indicates the phase When closing image block without residual error, it is determined that when performing inter prediction to the enhancement layer image block, the GRP algorithms are not used；Or, When the corresponding residual information of associated picture block includes the coded block flag of one or more transform blocks of the associated picture block, and One or more transform blocks of the associated picture block (can be the one or more transform blocks or any one specified Or multiple transform blocks) coded block flag when indicating that corresponding transform block does not include residual error non-zero transform coefficient, it is determined that to this When enhancement layer image block performs inter prediction, the GRP algorithms are not used；Or, in the corresponding residual information of associated picture block One or more transform blocks including the associated picture block (can be one or more transform blocks for specifying or any One or more transform blocks) include residual error non-zero transform coefficient when, and one or more transform blocks of the associated picture block The amplitude of residual error non-zero transform coefficient is less than the residual error non-zero of first threshold and one or more transform blocks of the associated picture block When the number of conversion coefficient is met less than at least one of Second Threshold, it is determined that performing frame to the enhancement layer image block Between when predicting, the GRP algorithms are not used.If being unsatisfactory for above-mentioned three kinds of conditions, when carrying out inter prediction, it can use GRP algorithms.

Signified first threshold can be according to the statistical true of residual error non-zero transform coefficient amplitude in embodiments of the present invention It is fixed, for example, the average value of the residual error non-zero transform coefficient amplitude of statistics is defined as into the first threshold.Signified Second Threshold can Determined with the statistical property of the residual error non-zero transform coefficient number according to transform block, for example, can be by a transform block of statistics The average value of the number for the residual error non-zero transform coefficient being had is defined as the Second Threshold.For different types of transform block, The value of threshold value can be different.For example, be a for red poor transform block first threshold value, Second Threshold value is b, for example, A can be the average value of the residual error non-zero transform coefficient amplitude of the poor transform block of red of statistics, and b can be the red counted The average value for the residual error non-zero transform coefficient number that poor transform block has；And for blue poor transform block, first threshold can be with Value is c, and Second Threshold can be using value as d, for example, c can be the residual error non-zero transform coefficient of the poor transform block of blueness of statistics The average value of amplitude, residual error non-zero transform coefficient number that d can have for a poor transform block of blueness of statistics is averaged Value.Also, above-mentioned threshold value can also be related to quantization step.

In order to be more clearly understood from the present invention, included separately below with the associated picture block of enhancement layer image block corresponding The associated picture block of Primary layer image block and enhancement layer image block includes illustrating exemplified by adjacent encoded image block.Its In, the coded block flag of image block is illustrates exemplified by rqt_root_cbf, but the embodiment of the present invention is not limited this It is fixed.

Hereafter illustrated first so that the associated picture block of enhancement layer image block includes corresponding Primary layer image block as an example.

In embodiments of the present invention, that enhancement layer image block can be obtained from basic layer bit stream is corresponding basic for code device The rqt_root_cbf of tomographic image block；When after the rqt_root_cbf indicates Primary layer image block coding without residual error, for example, Rqt_root_cbf indicates without residual error that code device can directly determine performing inter prediction to the enhancement layer image block for 0 When, GRP algorithms are not used.When the rqt_root_cbf indicates to have residual error after Primary layer image block coding, for example, rqt_ Root_cbf is that 1 instruction has residual error, and code device can directly determine, when performing inter prediction to the enhancement layer image block, to adopt Use GRP algorithms.

In embodiments of the present invention, though the rqt_root_cbf of the corresponding Primary layer image block of enhancement layer image block indicates to be somebody's turn to do There is residual error after Primary layer image block coding, code device can also further determine that a transform block of the Primary layer image block (can be any one transform block or specify a transform block) or multiple transform blocks (can be any number of conversion Block or the multiple transform blocks specified) whether include residual error non-zero transform coefficient；Specifically, code device can be from base The coded block flag of one or more transform blocks of the Primary layer image block is obtained in this layer bit stream, so that according to the Primary layer figure As the coded block flag of one or more transform blocks of block, it is determined that whether corresponding transform block includes residual error non-zero transform coefficient Number；Indicate that corresponding transform block does not include residual error in the coded block flag of one or more transform blocks of the Primary layer image block (wherein, if what is detected is the coded block flag of multiple transform blocks, the plurality of transform block is needed during non-zero transform coefficient not Include residual error non-zero transform coefficient), it can directly determine, when performing inter prediction to the enhancement layer image block, not use GRP Algorithm.The coded block flag of any one in one or more transform blocks of the Primary layer image block indicates corresponding conversion When block includes residual error non-zero transform coefficient, then it can directly determine, when performing inter prediction to the enhancement layer image block, to use GRP algorithms.

In embodiments of the present invention, in one or more transform blocks of the corresponding Primary layer image block of enhancement layer image block The coded block flag of some or multiple transform blocks when indicating that corresponding transform block includes residual error non-zero transform coefficient, coding dress Putting can also further determine that this some or multiple transform blocks number and amplitude of the residual error non-zero transform coefficient that include.Such as Really this some or multiple transform blocks in the amplitude of residual error non-zero transform coefficient that includes of each transform block be more than the first threshold Value and residual error non-zero transform coefficient number be more than Second Threshold when, then can determine use GRP algorithms, otherwise can determine not Using GRP algorithms.It is of course also possible to determine whether to use GRP algorithms in such a way：If this some or multiple conversion The amplitude for the residual error non-zero transform coefficient that each transform block in block includes is less than first threshold and residual error non-zero transform coefficient Number be less than Second Threshold, then can determine not using GRP algorithms, otherwise can determine to use GRP algorithms.

Secondly adjacent image block is hereafter included with the associated picture block of enhancement layer image block, and the adjacent image block is Illustrated exemplified by left image block and upper images block, it should be understood that the embodiment of the present invention is not limited to this.

In embodiments of the present invention, code device can obtain the upper images of enhancement layer image block from enhancement layer bitstream The rqt_root_cbf of the block and rqt_root_cbf of left image block；In the rqt_root_cbf instructions of the upper images block When the rqt_root_cbf without residual error and left image block is indicated after left image block coding without residual error after square image block coding, example Such as, rqt_root_cbf is indicated without residual error for 0, and code device can be determined directly pre- to enhancement layer image block execution interframe During survey, GRP algorithms are not used.There are residual error or a left side after the rqt_root_cbf of the upper images block indicates upper images block coding When the rqt_root_cbf of square image block indicates that image block coding in left has residual error, for example, rqt_root_cbf, which is 1 instruction, residual Difference, code device can be determined directly when performing inter prediction to the enhancement layer image block, using GRP algorithms.It is assumed hereinafter that There is residual error after upper images block and left image block coding.

In embodiments of the present invention, up the rqt_root_cbf of image block has residual error after indicating upper images block coding When indicating that image block coding in left has residual error with the rqt_root_cbf of left image block, code device can also be further determined that One or more transform blocks of the upper images block (can be arbitrary one or more transform blocks or specify one Individual or multiple transform blocks) and left image block one or more transform blocks (can be arbitrary one or more transform blocks, Can be the one or more transform blocks specified) whether include residual error non-zero transform coefficient；Specifically, code device can be from increasing The coded block flag of one or more transform blocks of the upper images block is obtained in strong layer bit stream, according to being somebody's turn to do for the upper images block The coded block flag of one or more transform blocks, determines whether the transform block of upper images block includes residual error non-zero transform coefficient, And the coded block flag of one or more transform blocks of the left image block is obtained from code stream, according to being somebody's turn to do for the left image block The coded block flag of one or more transform blocks, determines whether the transform block of left image block includes residual error non-zero transform coefficient； Indicate that corresponding transform block does not include residual error non-zero in the coded block flag of one or more transform blocks of the upper images block The coded block flag of one or more transform blocks of conversion coefficient and the upper images block indicates that corresponding transform block does not include During residual error non-zero transform coefficient, it can directly determine, when performing inter prediction to the enhancement layer image block, not use GRP to calculate Method.Indicate that corresponding transform block includes residual error non-zero in the coded block flag of one or more transform blocks of the upper images block The coded block flag of one or more transform blocks of conversion coefficient or the left image block indicates corresponding transform block comprising residual During poor non-zero transform coefficient, can directly it determine when performing inter prediction to the enhancement layer image block, using GRP algorithms.With Some lower assumed in only one or more transform blocks of left image block or the coded block flag of multiple transform blocks are indicated Corresponding transform block includes residual error non-zero transform coefficient.

In embodiments of the present invention, though certain in one or more transform blocks of the left image block of enhancement layer image block When the coded block flag of one or more transform blocks indicates that corresponding transform block includes residual error non-zero transform coefficient, code device is also May further determine that left image block this some or the number of residual error non-zero transform coefficient that includes of multiple transform blocks And amplitude.If including residual error non-zero transform coefficient amplitude be more than first threshold and residual error non-zero transform coefficient number it is big When Second Threshold, then it can determine to use GRP algorithms, otherwise, it determines not using GRP algorithms.It is of course also possible to according to following Mode determines whether to use GRP algorithms：The residual error non-zero that each transform block at this in some or multiple transform blocks includes When the amplitude of conversion coefficient is less than the number of first threshold and residual error non-zero transform coefficient less than Second Threshold, then it can determine not Using GRP algorithms, it otherwise can determine to use GRP algorithms.

It should be understood that in embodiments of the present invention, code device can also obtain enhancement layer image block directly from code stream The coded block flag of the transform block of associated picture block, rather than it is determined that the rqt_root_ of the associated picture block of enhancement layer image block Cbf has after residual error after indicating associated picture block coding, just obtains the transform block of the associated picture block of the enhancement layer image block Coded block flag.It should also be understood that in embodiments of the present invention, code device directly can also obtain enhancement layer image from code stream The residual error non-zero transform coefficient of one or more transform blocks of the associated picture block of block, and judge one or more transform blocks The size and amplitude of residual error non-zero transform coefficient, or it is determined that the rqt_root_cbf of the associated picture block of the enhancement layer image block Have after indicating associated picture block coding after residual error, one of the associated picture block of enhancement layer image block is obtained directly from code stream Or the residual error non-zero transform coefficient of multiple transform blocks, and judge one or more transform blocks residual error non-zero transform coefficient it is big Small and amplitude, rather than in the coded block flag for the transform block for obtaining the associated picture block, it is determined that corresponding transform block includes residual error The residual error non-zero that one or more transform blocks of the associated picture block of enhancement layer image block are just obtained after non-zero transform coefficient becomes Change coefficient.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag Include the corresponding Primary layer image block of the enhancement layer image block or the encoded image block adjacent with the enhancement layer image block, and according to The corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, if calculated using GRP Method, wherein, the corresponding residual information of associated picture block is used to indicate whether the associated picture block has residual error or the associated picture The residual error size degree of block, so that, only in the residual information according to associated picture block, it is determined that when needing to use GRP algorithms, Inter prediction is performed to enhancement layer image block using GRP algorithms, so as to be not affected substantially in guarantee picture quality In the case of, reduce video encoding complexity and reduce video encoding time.

The method for layered video coding according to embodiments of the present invention is described with reference to Fig. 2 and Fig. 3 above, below Fig. 4 will be combined and describe the method that layered video is decoded that is used for according to embodiments of the present invention.

Fig. 4 is the indicative flowchart of the method 200 for being used for layered video decoding according to embodiments of the present invention.Such as Fig. 4 Shown, this method 200 includes：

S210, determines the associated picture block of enhancement layer image block, and it is corresponding that the associated picture block includes enhancement layer image block Primary layer image block or the decoded image blocks adjacent with the enhancement layer image block；

S220, obtains the corresponding residual information of enhancement layer image block, according to the corresponding residual information of associated picture block, It is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, if GRP algorithms are used, in order to right The enhancement layer image block performs the motion compensation to obtain the prediction signal of the enhancement layer image block, and increased according to described The prediction signal of strong tomographic image block is decoded to the enhancement layer image block, wherein, the corresponding residual error letter of the associated picture block Cease for indicating whether the associated picture block has the residual error size degree of residual error or the associated picture block after the decoding after the decoding.

In SHVC, due in motion compensation, though code rate gains maximum in SHVC can be contribute to using GRP algorithms, But the complexity of decoding is also considerably increased, the time of decoding is extended, it is therefore desirable to it is determined that being calculated under what circumstances using GRP Method and which kind of in the case of without using GRP algorithms, can ensure picture quality while, reduce decoding complexity, reduction solution The time of code.

Specifically, decoding apparatus is being needed to the motion compensation in current EL image block perform decoding reconstruction operation When, it may be determined that the associated picture block of the enhancement layer image block, wherein, the associated picture block can include the decoding enhancement layer figure As the corresponding Primary layer image block of block or the decoded image blocks adjacent with the enhancement layer image block；Then, decoding apparatus can be with The corresponding residual information of associated picture block is obtained, wherein, the corresponding residual information of associated picture block is used to indicate the correlation Whether image block has the residual error size degree of residual error or the associated picture block after the decoding after the decoding；Getting correlation figure As block residual information after, decoding apparatus can be according to the residual information of the associated picture block, it is determined that to the enhancement layer image During motion compensation in block perform decoding reconstruction operation, if use GRP algorithms；When it is determined that needing to use GRP algorithms, then exist During to motion compensation in the enhancement layer image block perform decoding reconstruction operation, use GRP algorithms to obtain the enhancement layer image The prediction signal of block when it is determined that GRP algorithms need not be used, can be used with obtaining the prediction signal of the enhancement layer image block The movement compensating algorithm of single-layer video compress technique, i.e., in the motion compensation in carrying out decoding and rebuilding operation, it is not necessary to use Base layer data amendment enhancement layer prediction signal.Then, decoding apparatus can be according to the prediction signal, to the enhancement layer image block Perform decoding, for example, adding the residual error of the enhancement layer image block by the prediction signal of the enhancement layer image block, obtains reconstruct image As block.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag The corresponding Primary layer image block of enhancement layer image block or the decoded image blocks adjacent with the enhancement layer image block are included, according to the phase The corresponding residual information of image block is closed, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, Whether GRP algorithms are used, wherein, the corresponding residual information of associated picture block is used to indicate that the associated picture block is after the decoding It is no to have the residual error size degree of residual error or the associated picture block after the decoding, so that, only in the residual error according to associated picture block Information, it is determined that when needing to use GRP algorithms, just using GRP algorithms to the fortune in enhancement layer image block perform decoding reconstruction operation Dynamic compensation, so as in the case where ensureing that picture quality is not affected substantially, reduce video decoding complexity and subtract Few Capture Program Time.

In embodiments of the present invention, with enhancement layer image block adjacent decoded image blocks may be referred to shown in Fig. 2 with The adjacent encoded image block of enhancement layer image block, i.e. the upper left-hand image block including the enhancement layer image block, upper images block, Upper right side image block, left image block and lower left image block.Certainly, the decoded image blocks adjacent with enhancement layer image block Can be other image blocks, for example, upper left-hand image block, upper images block, upper right side image block, right image block and lower right Image block；Depending on specifically can be according to decoding order.The associated picture block of the enhancement layer image block can include and the enhancement layer The decoded image blocks of image block adjacent two or more (for example, upper left-hand image block and upper images blocks).Certainly, A decoded image blocks can also be only included.The embodiment of the present invention is defined not to this.

In embodiments of the present invention, the corresponding residual information of associated picture block of enhancement layer image block can include：The phase Close image block coded block flag (coding block flag, rqt_root_cbf), the associated picture block it is one or more In the residual error non-zero transform coefficient that one or more transform blocks of the coded block flag of transform block and the associated picture block include At least one, the coded block flag of the associated picture block is used for whether having residual error after indicating associated picture block decoding, the correlation The coded block flag of one or more transform blocks of image block is used for whether having residual error after indicating corresponding transform block decoding.Its In, the coded block flag of the image block can be residual error quaternary tree root coded block flag (the residual quad of the image block Tree root coding block flag, rqt_root_cbf), for example, rqt_root_cbf represents the associated picture for 0 Without residual error after block decoding, rqt_root_cbf has residual error after representing associated picture block decoding for 1.One of the associated picture block It is blue or the coded block flag of multiple transform blocks refers to the coded block flag (cbf_luma) of the luminance transformation block of the associated picture block One or many in the coded block flag (cbf_cb) of aberration transform block and the coded block flag (cbf_cr) of red poor transform block It is individual.Cbf_luma is used to indicate whether the luminance transformation block includes residual error non-zero transform coefficient, for example, cbf_luma is 1 expression The luminance transformation block includes residual error non-zero transform coefficient, and cbf_luma is that 0 expression luminance transformation block becomes not comprising residual error non-zero Change coefficient.Cbf_cr is used to indicate whether red poor transform block includes residual error non-zero transform coefficient, for example, cbf_cr represents for 1 Red difference transform block includes residual error non-zero transform coefficient, and cbf_cr is that the 0 poor transform block of the expression red does not include residual error non-zero Conversion coefficient.Cbf_cb is used to indicate whether the poor transform block of the blueness includes residual error non-zero transform coefficient, for example, cbf_cb is 1 Represent that the poor transform block of the blueness includes residual error non-zero transform coefficient, cbf_cb is that the 0 poor transform block of the expression blueness does not include residual error Non-zero transform coefficient.

In embodiments of the present invention, it is determined that during using GRP algorithms, decoding apparatus parses GRP marks from enhancement layer bitstream Will position, when it is determined that not using GRP algorithms, decoding apparatus parses GRP flag bits not from enhancement layer bitstream, so as to reduce The difficulty of decoding.

In an embodiment of the present invention, according to the corresponding residual information of associated picture block in S220, it is determined that right During motion compensation in the enhancement layer image block perform decoding reconstruction operation, if use GRP algorithms, can include：In the phase When closing coded block flag of the corresponding residual information of image block including the associated picture block, and the encoding block mark of the associated picture block When will indicates that the associated picture block has residual error after the decoding, it is determined that in the enhancement layer image block perform decoding reconstruction operation During motion compensation, using the GRP algorithms；Or, include the associated picture block in the corresponding residual information of associated picture block One or more transform blocks (can be the one or more transform blocks or any one or more transform blocks specified) During coded block flag, and the coded block flag of one or more transform blocks of the associated picture block indicates that corresponding transform block is included During residual error non-zero transform coefficient, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, using The GRP algorithms；Or, include one or more transform blocks of the associated picture block in the corresponding residual information of associated picture block Including residual error non-zero transform coefficient when, and the associated picture block one or more transform blocks (can be specify one or Multiple transform blocks or any one or more transform blocks) residual error non-zero transform coefficient amplitude be more than first threshold It is more than with the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block in Second Threshold at least When one kind is met, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, using the GRP Algorithm.When above-mentioned three kinds of conditions are not being met, then it can determine not using the GRP algorithms.

In another embodiment of the invention, according to the corresponding residual information of associated picture block in S220, it is determined that During to motion compensation in the enhancement layer image block perform decoding reconstruction operation, if use GRP algorithms, can include：At this When the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, and the encoding block of the associated picture block When mark indicates the associated picture block after the decoding without residual error, it is determined that in the enhancement layer image block perform decoding reconstruction operation Motion compensation when, the GRP algorithms are not used；Or, include the associated picture in the corresponding residual information of associated picture block One or more transform blocks of block (can be the one or more transform blocks or any one or more conversion specified Block) coded block flag when, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding conversion When block does not include residual error non-zero transform coefficient, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, the GRP algorithms are not used；Or, at one of the corresponding residual information of associated picture block including the associated picture block or During the residual error non-zero transform coefficient that multiple transform blocks include, and one or more transform blocks of the associated picture block (can refer to Fixed one or more transform blocks or any one or more transform blocks) residual error non-zero transform coefficient amplitude it is small It is less than the second threshold in first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of value is met, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, the GRP algorithms are not used.When above-mentioned three kinds of conditions are not being met, then it can determine to use the GRP algorithms.

In embodiments of the present invention, above-mentioned first threshold can be according to the statistical true of residual error non-zero transform coefficient amplitude It is fixed, for example, the average value of the residual error non-zero transform coefficient amplitude of statistics is defined as into the first threshold.The Second Threshold can root Determined according to the statistical property of the residual error non-zero transform coefficient number of transform block, for example, can be had a transform block of statistics The average value of the number of some residual error non-zero transform coefficients is defined as the Second Threshold.For different types of transform block, threshold value Value can be different.For example, being a for red poor transform block first threshold value, Second Threshold value is b, for example, a can Think the average value of the residual error non-zero transform coefficient amplitude of the poor transform block of red of statistics, b can be poor for a red of statistics The average value for the residual error non-zero transform coefficient number that transform block has；And for blue poor transform block, first threshold can take It is worth for c, Second Threshold can be using value as d.For example, c can be the residual error non-zero transform coefficient width of the poor transform block of blueness of statistics The average value of degree, the average value for the residual error non-zero transform coefficient number that d can have for the poor transform block of a blueness of statistics. Also, above-mentioned threshold value can also be related to quantization step.

In order to be more clearly understood from the present invention, included separately below with the associated picture block of enhancement layer image block corresponding The associated picture block of Primary layer image block and enhancement layer image block includes illustrating exemplified by adjacent decoded image blocks.Its In, the coded block flag of image block is illustrates exemplified by rqt_root_cbf, but the embodiment of the present invention is not limited this It is fixed.

Hereafter enter first so that the associated picture block of enhancement layer image block includes the corresponding Primary layer image block of enhancement layer as an example Row explanation.

In embodiments of the present invention, that enhancement layer image block can be obtained from basic layer bit stream is corresponding basic for decoding apparatus The rqt_root_cbf of tomographic image block；When after the rqt_root_cbf indicates Primary layer image block decoding without residual error, for example, Rqt_root_cbf indicates without residual error that decoding apparatus can directly determine to rebuild to the enhancement layer image block perform decoding for 0 During motion compensation in operation, GRP algorithms are not used.Have residual after the rqt_root_cbf indicates Primary layer image block decoding When poor, for example, rqt_root_cbf, which is 1 instruction, residual error, decoding apparatus can directly determine to hold to the enhancement layer image block Row decoding and rebuilding operation in motion compensation when, using GRP algorithms.

In embodiments of the present invention, being indicated in the rqt_root_cbf of the corresponding Primary layer image block of enhancement layer image block should When having residual error after the decoding of Primary layer image block, decoding apparatus can also further determine that a transform block of the Primary layer image block (can be any one transform block or specify a transform block) or multiple transform blocks (can be any number of conversion Block or the multiple transform blocks specified) whether include residual error non-zero transform coefficient；Specifically, decoding apparatus can be obtained The coded block flag of one or more transform blocks of the Primary layer image block, thus according to this of the Primary layer image block or The coded block flag of multiple transform blocks, it is determined that whether corresponding transform block includes residual error non-zero transform coefficient；In the Primary layer figure When indicating that corresponding transform block does not include residual error non-zero transform coefficient as the coded block flag of one or more transform blocks of block (wherein, if what is detected is the coded block flag of multiple transform blocks, need the plurality of transform block not include residual error non-zero and become Change coefficient), it can directly determine, in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, not use GRP algorithms.Any one coded block flag in one or more transform blocks of the Primary layer image block indicates corresponding become When changing block comprising residual error non-zero transform coefficient, then can directly it determine in the enhancement layer image block perform decoding reconstruction operation Motion compensation when, using GRP algorithms.

In embodiments of the present invention, in one or more transform blocks of the corresponding Primary layer image block of enhancement layer image block The coded block flag of some or multiple transform blocks when indicating that corresponding transform block includes residual error non-zero transform coefficient, decoding dress Putting can also further determine that this some or multiple transform blocks number and amplitude of the residual error non-zero transform coefficient that include.Such as Really this some or multiple transform blocks in the amplitude of residual error non-zero transform coefficient that includes of each transform block be more than the first threshold Value and residual error non-zero transform coefficient number be more than Second Threshold when, then can determine use GRP algorithms, otherwise can be defined as GRP algorithms are not used.It can also determine whether to use GRP algorithms in such a way：At this in some or multiple transform blocks The amplitude of residual error non-zero transform coefficient that includes of each transform block be less than of first threshold and residual error non-zero transform coefficient When number is less than Second Threshold, then it can determine not using GRP algorithms, otherwise using GRP algorithms.

Secondly adjacent decoded image blocks are hereafter included with the associated picture block of enhancement layer image block, and this is adjacent Decoded image blocks is illustrate exemplified by left image block and upper images block, it should be understood that the embodiment of the present invention is not limited to This.

In embodiments of the present invention, decoding apparatus can obtain the upper images of enhancement layer image block from enhancement layer bitstream The rqt_root_cbf of the block and rqt_root_cbf of left image block；In the rqt_root_cbf instructions of the upper images block When the rqt_root_cbf without residual error and left image block is indicated after the decoding of left image block without residual error after square image block decoding, example Such as, rqt_root_cbf indicates that without residual error decoding apparatus can be determined directly to the enhancement layer image block perform decoding weight for 0 When building the motion compensation in operation, GRP algorithms are not used.Upper images block is indicated in the rqt_root_cbf of the upper images block When the rqt_root_cbf for having residual error or left image block after decoding indicates to have residual error after the decoding of left image block, for example, rqt_ Root_cbf is that 1 instruction has residual error, and decoding apparatus can be determined directly to the enhancement layer image block perform decoding reconstruction operation In motion compensation when, using GRP algorithms.But it is assumed hereinafter that there is residual error after upper images block and the decoding of left image block.

In embodiments of the present invention, up the rqt_root_cbf of image block has residual error after indicating the decoding of upper images block There is residual error after indicating the decoding of left image block with the rqt_root_cbf of left image block, decoding apparatus can also be further determined that One or more transform blocks of the upper images block (can be arbitrary one or more transform blocks or specify one Individual or multiple transform blocks) and left image block one or more transform blocks (can be arbitrary one or more transform blocks, Can be the one or more transform blocks specified) whether include residual error non-zero transform coefficient；Specifically, decoding apparatus can be from increasing The coded block flag of one or more transform blocks of the upper images block is obtained in strong layer bit stream, according to being somebody's turn to do for the upper images block The coded block flag of one or more transform blocks, determines whether the transform block of upper images block includes residual error non-zero transform coefficient, And the coded block flag of one or more transform blocks of the left image block is obtained from enhancement layer bitstream, according to the left image The coded block flag of one or more transform blocks of block, determines whether the transform block of left image block includes residual error non-zero transform Coefficient；Indicate that corresponding transform block does not include residual error in the coded block flag of one or more transform blocks of the upper images block The coded block flag of one or more transform blocks of non-zero transform coefficient and the upper images block indicates corresponding transform block not During comprising residual error non-zero transform coefficient, the motion in the enhancement layer image block perform decoding reconstruction operation can be directly determined During compensation, GRP algorithms are not used.Indicated in the coded block flag of one or more transform blocks of the upper images block corresponding The coded block flag of the one or more transform blocks of the transform block comprising residual error non-zero transform coefficient or the left image block is indicated When corresponding transform block includes residual error non-zero transform coefficient, it can directly determine to rebuild to the enhancement layer image block perform decoding During motion compensation in operation, using GRP algorithms.It is assumed hereinafter that only in one or more transform blocks of left image block The coded block flag of some or multiple transform blocks indicates that corresponding transform block includes residual error non-zero transform coefficient.

In embodiments of the present invention, it is a certain in one or more transform blocks of the left image block of enhancement layer image block When the coded block flag of individual or multiple transform blocks indicates that corresponding transform block includes residual error non-zero transform coefficient, decoding apparatus may be used also With further determine that left image block this some or the number of residual error non-zero transform coefficient that includes of multiple transform blocks and Amplitude.If this some or multiple transform blocks in the amplitude of residual error non-zero transform coefficient that includes of each transform block be more than When first threshold and the number of residual error non-zero transform coefficient are more than Second Threshold, then it can determine to use GRP algorithms, otherwise can be with It is defined as not using GRP algorithms.It can also determine whether to use GRP algorithms in such a way：At this some or it is multiple become The amplitude for the residual error non-zero transform coefficient that each transform block changed in block includes is less than first threshold and residual error non-zero transform coefficient When several numbers is less than Second Threshold, then it can determine not using GRP algorithms, otherwise using GRP algorithms.

It should be understood that in embodiments of the present invention, decoding apparatus can also obtain enhancement layer image block directly from code stream The coded block flag of the transform block of associated picture block, rather than it is determined that the rqt_root_ of the associated picture block of enhancement layer image block Cbf has after residual error after indicating associated picture block decoding, just obtains the transform block of the associated picture block of the enhancement layer image block Coded block flag.It should also be understood that in embodiments of the present invention, decoding apparatus directly can also obtain enhancement layer image from code stream The residual error non-zero transform coefficient of one or more transform blocks of the associated picture block of block, and judge one or more transform blocks The size and amplitude of residual error non-zero transform coefficient, or it is determined that the rqt_root_cbf of the associated picture block of the enhancement layer image block Have after indicating associated picture block decoding after residual error, one of the associated picture block of enhancement layer image block is obtained directly from code stream Or the residual error non-zero transform coefficient of multiple transform blocks, and judge one or more transform blocks residual error non-zero transform coefficient it is big Small and amplitude, rather than in the coded block flag for the transform block for obtaining the associated picture block, it is determined that corresponding transform block includes residual error The residual error non-zero that one or more transform blocks of the associated picture block of enhancement layer image block are just obtained after non-zero transform coefficient becomes Change coefficient.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag Include the corresponding Primary layer image block of the enhancement layer image block or the decoded image blocks adjacent with the enhancement layer image block, and according to The corresponding residual information of associated picture block, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, if GRP algorithms are used, wherein, the corresponding residual information of associated picture block is used to indicate the associated picture block in decoding Whether there is the residual error size degree of residual error or the associated picture block after the decoding afterwards, so that, only according to associated picture block Residual information, it is determined that when needing to use GRP algorithms, just using GRP algorithms in enhancement layer image block perform decoding reconstruction operation Motion compensation, so as to which in the case where ensureing that picture quality is not affected substantially, reduction decoding complex degree simultaneously subtracts The time is decoded less.

The method for layered video coding according to embodiments of the present invention is described with reference to Fig. 1 to Fig. 4 above to conciliate The method of code.Encoding apparatus and decoding apparatus according to embodiments of the present invention are described below with reference to Fig. 5 to Figure 10.

Fig. 5 is the schematic block diagram of code device 300 according to embodiments of the present invention.The code device 300 can be compiled Code device.As shown in figure 5, the code device 300 includes：

First determining unit 310, the associated picture block for determining enhancement layer image block, associated picture block includes enhancing The corresponding Primary layer image block of tomographic image block or the encoded image block adjacent with the enhancement layer image block；

Acquiring unit 340, for obtaining the corresponding residual information of associated picture block；

Second determining unit 320, for according to the corresponding residual information of associated picture block, it is determined that to the enhancement layer figure When performing inter prediction as block, if use vague generalization residual prediction algorithm, in order to perform institute to the enhancement layer image block Inter prediction is stated to obtain the prediction signal of the enhancement layer image block, and according to the prediction signal pair of the enhancement layer image block The enhancement layer image block is encoded, wherein, the corresponding residual information of associated picture block is used to indicate the associated picture block Whether residual error or the residual error size degree of the associated picture block are had.

Alternatively, the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, the correlation It is residual that the coded block flag of one or more transform blocks of image block and one or more transform blocks of the associated picture block include At least one of poor non-zero transform coefficient, wherein, the coded block flag of the associated picture block is used to indicate the related figure As whether block has a residual error, the coded block flag of one or more transform blocks of the associated picture block is used to indicate corresponding conversion Whether block has residual error.

Alternatively, second determining unit 320 specifically for：

Alternatively, the associated picture block includes the two or more encoded figures adjacent with the enhancement layer image block As block.

Alternatively, as shown in fig. 6, the code device 300 also includes writing unit 330, in second determining unit When 320 determinations use the vague generalization residual prediction algorithm, vague generalization residual prediction flag position is write in enhancement layer bitstream, its In, the vague generalization residual prediction flag position write in the enhancement layer bitstream is set to indicate that the vague generalization residual prediction The pattern of algorithm, when second determining unit 320 determines not use the vague generalization residual prediction algorithm, not by vague generalization residual error In prediction indication position write-in enhancement layer bitstream.

It is layered it should be understood that code device 300 according to embodiments of the present invention may correspond to being used in the embodiment of the present invention Above and other operation of unit in code device in the method for Video coding, and code device 300 and/or work( The corresponding flow of the method 100 in Fig. 1 can be realized respectively, for sake of simplicity, will not be repeated here.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag Include the corresponding Primary layer image block of the enhancement layer image block or the encoded image block adjacent with the enhancement layer image block, and according to The corresponding residual information of associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, if calculated using GRP Method, wherein, the corresponding residual information of associated picture block is used to indicate whether the associated picture block has residual error or the associated picture The residual error size degree of block, so that, only in the residual information according to associated picture block, it is determined that when needing to use GRP algorithms, Inter prediction is performed to enhancement layer image block using GRP algorithms, so as to be not affected substantially in guarantee picture quality In the case of, reduce encoder complexity and reduce the scramble time.

Fig. 7 is the schematic block diagram of decoding apparatus 400 according to embodiments of the present invention.The decoding apparatus 400 can be solution Code device.As shown in fig. 7, the decoding apparatus 400 includes：

First determining unit 410, the associated picture block for determining enhancement layer image block, associated picture block includes enhancing The corresponding Primary layer image block of tomographic image block or the decoded image blocks adjacent with the enhancement layer image block；

Acquiring unit 440, for obtaining the corresponding residual information of enhancement layer image block；

Second determining unit 420, for according to the corresponding residual information of associated picture block, it is determined that to the enhancement layer figure During as motion compensation in block perform decoding reconstruction operation, if vague generalization residual prediction algorithm is used, in order to increase to described Strong tomographic image block performs the motion compensation to obtain the prediction signal of the enhancement layer image block, and according to the enhancement layer figure As the prediction signal of block is decoded to the enhancement layer image block, wherein, the corresponding residual information of associated picture block is used for Indicate whether the associated picture block has the residual error size degree of residual error or the associated picture block after the decoding after the decoding.

Alternatively, the corresponding residual information of associated picture block includes the coded block flag of the associated picture block, the correlation It is residual that the coded block flag of one or more transform blocks of image block and one or more transform blocks of the associated picture block include At least one of poor non-zero transform coefficient.

Alternatively, second determining unit 420 specifically for：

Alternatively, the associated picture block includes having decoded figure with the two neighboring or more than two of the enhancement layer image block As block.

Alternatively, as shown in figure 8, the decoding apparatus 400 also includes resolution unit 430, in second determining unit When 420 determinations use the vague generalization residual prediction algorithm, vague generalization residual prediction flag position is parsed from enhancement layer bitstream, with root The pattern of the vague generalization residual prediction algorithm is determined according to the setting of vague generalization residual prediction flag position；In second determination When unit 420 determines not use the vague generalization residual prediction algorithm, vague generalization residual prediction mark is parsed not from enhancement layer bitstream Will position.

It is layered it should be understood that decoding apparatus 400 according to embodiments of the present invention may correspond to being used in the embodiment of the present invention Above and other operation of unit in code device in the method for Video coding, and decoding apparatus 400 and/or work( The corresponding flow of the method 200 in Fig. 4 can be realized respectively, for sake of simplicity, will not be repeated here.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag Include the corresponding Primary layer image block of the enhancement layer image block or the decoded image blocks adjacent with the enhancement layer image block, and according to The corresponding residual information of associated picture block, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, if GRP algorithms are used, wherein, the corresponding residual information of associated picture block is used to indicate the associated picture block in decoding Whether there is the residual error size degree of residual error or the associated picture block after the decoding afterwards, so that, only according to associated picture block Residual information, it is determined that when needing to use GRP algorithms, just using GRP algorithms in enhancement layer image block perform decoding reconstruction operation Motion compensation, simultaneously dropped so as in the case where ensureing that picture quality is not affected substantially, reduce decoding complex degree The low decoding time.

Fig. 9 is the schematic block diagram of code device 500 according to embodiments of the present invention.The code device 500 can be compiled Code device.As shown in figure 9, the code device 500 includes memory 510 and processor 520.The store program codes of memory 510, And processor 520 can call the program code stored in memory 510 to perform following operate：

The associated picture block of enhancement layer image block is determined, it is corresponding basic that the associated picture block includes enhancement layer image block Tomographic image block or the encoded image block adjacent with the enhancement layer image block；

According to the corresponding residual information of the associated picture block, it is determined that performing inter prediction to the enhancement layer image block When, if vague generalization residual prediction algorithm is used, in order to be obtained to the enhancement layer image block execution inter prediction The prediction signal of the enhancement layer image block, and according to the prediction signal of the enhancement layer image block to the enhancement layer image block Encoded, wherein, the corresponding residual information of the associated picture block be used to indicating the associated picture block whether have residual error or The residual error size degree of the associated picture block.

Alternatively, the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, institute State the coded block flag of one or more transform blocks of associated picture block and one or more transform blocks of the associated picture block Including at least one of residual error non-zero transform coefficient, wherein, the coded block flag of the associated picture block is used to indicate institute State whether associated picture block has residual error, the coded block flag of one or more transform blocks of the associated picture block is used to indicate phase Whether the transform block answered has residual error.

Alternatively, processor 520 can call the program code stored in memory 510 specifically to perform following operate：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and it is described When the coded block flag of associated picture block indicates that the associated picture block has residual error, it is determined that being performed to the enhancement layer image block During inter prediction, using the vague generalization residual prediction algorithm；Or,

Include one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During coded block flag, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block bag During non-zero transform coefficient containing residual error, it is determined that when performing inter prediction to the enhancement layer image block, it is residual using the vague generalization Poor prediction algorithm；Or,

Include one or more transform block bags of the associated picture block in the corresponding residual information of the associated picture block During the residual error non-zero transform coefficient included, and the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block The number that amplitude is more than the residual error non-zero transform coefficient of one or more transform blocks of first threshold and the associated picture block is big When at least one of Second Threshold is met, it is determined that when performing inter prediction to the enhancement layer image block, using The vague generalization residual prediction algorithm.

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and it is described When associated picture block coded block flag indicates the associated picture block without residual error, it is determined that performing frame to the enhancement layer image block Between when predicting, the vague generalization residual prediction algorithm is not used；Or,

Include one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During coded block flag, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block not During comprising residual error non-zero transform coefficient, it is determined that when performing inter prediction to the enhancement layer image block, not using described general Change residual prediction algorithm；Or,

Include one or more transform block bags of the associated picture block in the corresponding residual information of the associated picture block During the residual error non-zero transform coefficient included, and the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block The number that amplitude is less than the residual error non-zero transform coefficient of one or more transform blocks of first threshold and the associated picture block is small When at least one of Second Threshold is met, it is determined that when performing inter prediction to the enhancement layer image block, not adopting With the vague generalization residual prediction algorithm.

Alternatively, the associated picture block includes adjacent with the enhancement layer image block two or more compiled Code image block.

Alternatively, processor 520 calls the program code stored in memory 510 to can also carry out following operation：

It is determined that during using the vague generalization residual prediction algorithm, by vague generalization residual prediction flag position write-in enhancement layer code In stream, wherein, the vague generalization residual prediction flag position write in the enhancement layer bitstream is set to indicate that the vague generalization The pattern of residual prediction algorithm；

When it is determined that not using the vague generalization residual prediction algorithm, the write-in of vague generalization residual prediction flag position is not strengthened In layer bit stream.

It is layered it should be understood that code device 500 according to embodiments of the present invention may correspond to being used in the embodiment of the present invention Above and other operation of unit in code device in the method for Video coding, and code device 500 and/or work( The corresponding flow of the method 100 in Fig. 1 can be realized respectively, for sake of simplicity, will not be repeated here.

Figure 10 is the schematic block diagram of decoding apparatus 600 according to embodiments of the present invention.The decoding apparatus 600 can be solution Code device.As shown in Figure 10, the code device 600 includes memory 610 and processor 620.The storage program generation of memory 610 Code, and processor 620 can call the program code stored in memory 610 to perform following operate：

The associated picture block of enhancement layer image block is determined, it is corresponding basic that the associated picture block includes enhancement layer image block Tomographic image block or the decoded image blocks adjacent with the enhancement layer image block；

According to the corresponding residual information of the associated picture block, it is determined that being rebuild to the enhancement layer image block perform decoding During motion compensation in operation, if use vague generalization residual prediction algorithm, in order to perform institute to the enhancement layer image block Motion compensation is stated to obtain the prediction signal of the enhancement layer image block, and according to the prediction signal pair of the enhancement layer image block The enhancement layer image block is decoded, wherein, the corresponding residual information of the associated picture block is used to indicate the related figure As whether block has the residual error size degree of residual error or the associated picture block after the decoding after the decoding.

Alternatively, the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, institute State the coded block flag of one or more transform blocks of associated picture block and one or more transform blocks of the associated picture block Including at least one of residual error non-zero transform coefficient.

Alternatively, processor 620 calls the program code stored in memory 610 specifically to perform following operate：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and it is described When the coded block flag of associated picture block indicates that the associated picture block has residual error after the decoding, it is determined that to the enhancement layer figure During as motion compensation in block perform decoding reconstruction operation, using the vague generalization residual prediction algorithm；Or,

Include one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During coded block flag, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block bag During non-zero transform coefficient containing residual error, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, Using the vague generalization residual prediction algorithm；Or,

Include one or more transform block bags of the associated picture block in the corresponding residual information of the associated picture block During the residual error non-zero transform coefficient included, and the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block The number that amplitude is more than the residual error non-zero transform coefficient of one or more transform blocks of first threshold and the associated picture block is big When at least one of Second Threshold is met, it is determined that in the enhancement layer image block perform decoding reconstruction operation During motion compensation, using the vague generalization residual prediction algorithm.

Alternatively, processor 620 can call the program code stored in memory 610 specifically to perform following operate：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and it is described When the coded block flag of associated picture block indicates the associated picture block after the decoding without residual error, it is determined that to the enhancement layer figure During as motion compensation in block perform decoding reconstruction operation, the vague generalization residual prediction algorithm is not used；Or,

Include one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During coded block flag, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block not During comprising residual error non-zero transform coefficient, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, the vague generalization residual prediction algorithm is not used；Or,

Include one or more transform block bags of the associated picture block in the corresponding residual information of the associated picture block During the residual error non-zero transform coefficient included, and the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block The number that amplitude is less than the residual error non-zero transform coefficient of one or more transform blocks of first threshold and the associated picture block is small When at least one of Second Threshold is met, it is determined that in the enhancement layer image block perform decoding reconstruction operation During motion compensation, the vague generalization residual prediction algorithm is not used；

Alternatively, the associated picture block includes having solved with the two neighboring or more than two of the enhancement layer image block Code image block.

Alternatively, processor 620 calls the program code stored in memory 610 to can also carry out following operation：

It is determined that during using the vague generalization residual prediction algorithm, vague generalization residual prediction mark is parsed from enhancement layer bitstream Will position, to determine the pattern of the vague generalization residual prediction algorithm according to the setting of vague generalization residual prediction flag position；

When it is determined that not using the vague generalization residual prediction algorithm, vague generalization residual error is parsed not from enhancement layer bitstream pre- Survey flag bit.

It is layered it should be understood that decoding apparatus 600 according to embodiments of the present invention may correspond to being used in the embodiment of the present invention Above and other operation of unit in code device in the method for Video coding, and decoding apparatus 600 and/or work( The corresponding flow of the method 200 in Fig. 4 can be realized respectively, for sake of simplicity, will not be repeated here.

Therefore, in embodiments of the present invention, by determining the associated picture block of enhancement layer image block, the associated picture block bag Include the corresponding Primary layer image block of the enhancement layer image block or the decoded image blocks adjacent with the enhancement layer image block, and according to The corresponding residual information of associated picture block, it is determined that the motion compensation in the enhancement layer image block perform decoding reconstruction operation When, if GRP algorithms are used, wherein, the corresponding residual information of associated picture block is used to indicate the associated picture block in decoding Whether there is the residual error size degree of residual error or the associated picture block after the decoding afterwards, so that, only according to associated picture block Residual information, it is determined that when needing to use GRP algorithms, motion compensation is just performed to enhancement layer image block using GRP algorithms, so as to In the case where ensureing that picture quality is not affected substantially, to reduce decoding complex degree and reduce the decoding time.

Figure 11 is the schematic block diagram of decoder 700 according to embodiments of the present invention.As shown in figure 11, the decoder 700 Including entropy decoding unit 710, predicting unit 720, coefficient scanning unit 730, inverse quantization unit 740, inverse transformation block 750, plus Musical instruments used in a Buddhist or Taoist mass 760 and frame storaging piece 770.

The bit stream that 710 pairs of entropy decoding unit is received is carried out after entropy decoding, obtains header information and 1 dimensional vector.

Predicting unit 720 carries out motion compensation according to header information, to obtain prediction block, wherein, to enhancement layer image Block is performed before motion compensation, it may be determined that whether is performed motion compensation using GRP patterns, specifically can first be determined the enhancement layer figure As block associated picture block, the associated picture block include the corresponding Primary layer image block of enhancement layer image block or with the enhancing The adjacent encoded image block of tomographic image block, then obtains the corresponding residual information of the associated picture block, according to the correlation The corresponding residual information of image block, it is determined that when performing inter prediction to the enhancement layer image block, if residual using vague generalization Poor prediction algorithm, in order to obtain the pre- of the enhancement layer image block to the enhancement layer image block execution inter prediction Signal is surveyed, and the enhancement layer image block is encoded according to the prediction signal of the enhancement layer image block, wherein, the phase Closing the corresponding residual information of image block is used to indicate whether the associated picture block has the residual error of residual error or the associated picture block Size degree.It is specific to judge whether to use GRP algorithms, it may be referred to be described above, will not be repeated here.Predicting unit 720 The feedback information of receiving frame storaging piece 770, if frame storaging piece 770 is without memory space, predicting unit 720 is stopped.

Coefficient scanning unit 730 is performed coefficient scanning, inverse quantization unit 740 successively and performs re-quantization, against just by 1 dimensional vector Converter unit 750 is handed over to obtain residual block after performing inverse orthogonal transformation.Adder 760 is obtained according to prediction block and the residual block rebuild To the image block of reconstruction, the image block of reconstruction is put into frame storaging piece 770, and shown at display.

Those of ordinary skill in the art are it is to be appreciated that the list of each example described with reference to the embodiments described herein Member and algorithm steps, can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually Performed with hardware or software mode, depending on the application-specific and design constraint of technical scheme.Professional and technical personnel Described function can be realized using distinct methods to each specific application, but this realization is it is not considered that exceed The scope of the present invention.

It is apparent to those skilled in the art that, for convenience and simplicity of description, the system of foregoing description, The specific work process of device and unit, may be referred to the corresponding process in preceding method embodiment, will not be repeated here.

, can be with several embodiments provided herein, it should be understood that disclosed systems, devices and methods Realize by another way.For example, device embodiment described above is only schematical, for example, the unit Divide, only a kind of division of logic function there can be other dividing mode when actually realizing, such as multiple units or component Another system can be combined or be desirably integrated into, or some features can be ignored, or do not perform.It is another, it is shown or The coupling each other discussed or direct-coupling or communication connection can be the indirect couplings of device or unit by some interfaces Close or communicate to connect, can be electrical, machinery or other forms.

The unit illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple On NE.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.

In addition, each functional unit in each embodiment of the invention can be integrated in a processing unit, can also That unit is individually physically present, can also two or more units it is integrated in a unit.

If the function is realized using in the form of SFU software functional unit and is used as independent production marketing or in use, can be with It is stored in a computer read/write memory medium.Understood based on such, technical scheme is substantially in other words The part contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter Calculation machine software product is stored in a storage medium, including some instructions are to cause a computer equipment (can be individual People's computer, server, or network equipment etc.) perform all or part of step of each of the invention embodiment methods described. And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.

The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.

Claims

1. a kind of method for layered video coding, it is characterised in that including：

The associated picture block of enhancement layer image block is determined, the associated picture block includes the corresponding Primary layer figure of enhancement layer image block As block or the encoded image block adjacent with the enhancement layer image block；

According to the corresponding residual information of the associated picture block, it is determined that when performing inter prediction to the enhancement layer image block, Whether vague generalization residual prediction algorithm is used, it is described to obtain in order to perform the inter prediction to the enhancement layer image block The prediction signal of enhancement layer image block, and the enhancement layer image block is carried out according to the prediction signal of the enhancement layer image block Coding, wherein, the corresponding residual information of the associated picture block is used to indicate whether the associated picture block has residual error or described The residual error size degree of associated picture block.

2. according to the method described in claim 1, it is characterised in that the corresponding residual information of the associated picture block includes described The coded block flag of associated picture block, the coded block flag of one or more transform blocks of the associated picture block to it is described related At least one of residual error non-zero transform coefficient that one or more transform blocks of image block include, wherein, the associated picture The coded block flag of block is used to indicate whether the associated picture block has residual error, one or more conversion of the associated picture block The coded block flag of block is used to indicate whether corresponding transform block has residual error.

3. method according to claim 2, it is characterised in that described to be believed according to the corresponding residual error of the associated picture block Breath, it is determined that when performing inter prediction to the enhancement layer image block, if use vague generalization residual prediction algorithm, including：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and the correlation When the coded block flag of image block indicates that the associated picture block has residual error, it is determined that performing interframe to the enhancement layer image block During prediction, using the vague generalization residual prediction algorithm；Or, include in the corresponding residual information of the associated picture block described During the coded block flag of one or more transform blocks of associated picture block, and one or more transform blocks of the associated picture block Coded block flag when indicating that corresponding transform block includes residual error non-zero transform coefficient, it is determined that being held to the enhancement layer image block During row inter prediction, using the vague generalization residual prediction algorithm；Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of the associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block It is more than the more than first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of two threshold values are met, it is determined that when performing inter prediction to the enhancement layer image block, using described Vague generalization residual prediction algorithm.

4. method according to claim 2, it is characterised in that described to be believed according to the corresponding residual error of the associated picture block Breath, it is determined that when performing inter prediction to the enhancement layer image block, if use vague generalization residual prediction algorithm, including：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and the correlation When the coded block flag of image block indicates the associated picture block without residual error, it is determined that performing interframe to the enhancement layer image block During prediction, the vague generalization residual prediction algorithm is not used；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates that corresponding transform block does not include During residual error non-zero transform coefficient, it is determined that when performing inter prediction to the enhancement layer image block, not using the vague generalization residual Poor prediction algorithm；Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of the associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block It is less than the less than first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of two threshold values are met, it is determined that when performing inter prediction to the enhancement layer image block, institute is not used State vague generalization residual prediction algorithm.

5. method according to claim 4, it is characterised in that the associated picture block includes and the enhancement layer image block Adjacent two or more encoded image blocks.

6. method according to any one of claim 1 to 5, it is characterised in that methods described also includes：

It is determined that during using the vague generalization residual prediction algorithm, by vague generalization residual prediction flag position write-in enhancement layer bitstream In, wherein, the vague generalization residual prediction flag position write in the enhancement layer bitstream is set to indicate that the vague generalization is residual The pattern of poor prediction algorithm；

When it is determined that not using the vague generalization residual prediction algorithm, not by vague generalization residual prediction flag position write-in enhancement layer code In stream.

7. a kind of method decoded for layered video, it is characterised in that including：

The associated picture block of enhancement layer image block is determined, the associated picture block includes the corresponding Primary layer figure of enhancement layer image block As block or the decoded image blocks adjacent with the enhancement layer image block；

According to the corresponding residual information of the associated picture block, it is determined that to the enhancement layer image block perform decoding reconstruction operation In motion compensation when, if vague generalization residual prediction algorithm is used, in order to perform the fortune to the enhancement layer image block Dynamic compensation is to obtain the prediction signal of the enhancement layer image block, and according to the prediction signal of the enhancement layer image block to described Enhancement layer image block is decoded, wherein, the corresponding residual information of the associated picture block is used to indicate the associated picture block Whether residual error or the residual error size degree of the associated picture block are had.

8. method according to claim 7, it is characterised in that the corresponding residual information of the associated picture block includes described The coded block flag of associated picture block, the coded block flag of one or more transform blocks of the associated picture block to it is described related At least one of residual error non-zero transform coefficient that one or more transform blocks of image block include.

9. method according to claim 8, it is characterised in that described to be believed according to the corresponding residual error of the associated picture block Breath, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, if use vague generalization residual error Prediction algorithm, including：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and the correlation When the coded block flag of image block indicates that the associated picture block has residual error, it is determined that to the enhancement layer image block perform decoding During motion compensation in reconstruction operation, using the vague generalization residual prediction algorithm；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates corresponding transform block comprising residual During poor non-zero transform coefficient, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, using The vague generalization residual prediction algorithm；Or, include the associated picture block in the corresponding residual information of the associated picture block One or more transform blocks include residual error non-zero transform coefficient when, and one or more transform blocks of the associated picture block Residual error non-zero transform coefficient amplitude be more than first threshold and the associated picture block one or more transform blocks residual error When the number of non-zero transform coefficient is met more than at least one of Second Threshold, it is determined that to the enhancement layer image block During motion compensation in perform decoding reconstruction operation, using the vague generalization residual prediction algorithm.

10. method according to claim 8, it is characterised in that described to be believed according to the corresponding residual error of the associated picture block Breath, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, if use vague generalization residual error Prediction algorithm, including：

When the corresponding residual information of the associated picture block includes the coded block flag of the associated picture block, and the correlation When the coded block flag of image block indicates the associated picture block without residual error, it is determined that to the enhancement layer image block perform decoding During motion compensation in reconstruction operation, the vague generalization residual prediction algorithm is not used；Or,

Include the coding of one or more transform blocks of the associated picture block in the corresponding residual information of the associated picture block During block mark, and the coded block flag of one or more transform blocks of the associated picture block indicates that corresponding transform block does not include During residual error non-zero transform coefficient, it is determined that in the motion compensation in the enhancement layer image block perform decoding reconstruction operation, no Using the vague generalization residual prediction algorithm；Or,

What the one or more transform blocks for including the associated picture block in the corresponding residual information of the associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block It is less than the less than first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of two threshold values are met, it is determined that the motion in the enhancement layer image block perform decoding reconstruction operation During compensation, the vague generalization residual prediction algorithm is not used；

11. method according to claim 10, it is characterised in that the associated picture block includes and the enhancement layer image The two neighboring or more than two decoded image blocks of block.

12. the method according to any one of claim 7 to 11, it is characterised in that methods described also includes：

It is determined that during using the vague generalization residual prediction algorithm, vague generalization residual prediction flag is parsed from enhancement layer bitstream Position, to determine the pattern of the vague generalization residual prediction algorithm according to the setting of vague generalization residual prediction flag position；

When it is determined that not using the vague generalization residual prediction algorithm, vague generalization residual prediction mark is parsed not from enhancement layer bitstream Will position.

13. a kind of code device, it is characterised in that including：

First determining unit, the associated picture block for determining enhancement layer image block, the associated picture block includes enhancement layer figure As the corresponding Primary layer image block of block or the encoded image block adjacent with the enhancement layer image block；

Second determining unit, for according to the corresponding residual information of the associated picture block, it is determined that to the enhancement layer image It is described in order to be performed to the enhancement layer image block when block performs inter prediction, if use vague generalization residual prediction algorithm Inter prediction is to obtain the prediction signal of the enhancement layer image block, and according to the prediction signal of the enhancement layer image block to institute Enhancement layer image block is stated to be encoded, wherein, the corresponding residual information of the associated picture block is used to indicate the associated picture Whether block has residual error or the residual error size degree of the associated picture block.

14. code device according to claim 13, it is characterised in that the corresponding residual information bag of the associated picture block Include the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block and institute At least one of residual error non-zero transform coefficient that one or more transform blocks of associated picture block include is stated, wherein, the phase Closing the coded block flag of image block is used to indicate whether the associated picture block has residual error, one or many of the associated picture block The coded block flag of individual transform block is used to indicate whether corresponding transform block has residual error.

15. code device according to claim 14, it is characterised in that second determining unit specifically for：

16. code device according to claim 14, it is characterised in that second determining unit specifically for：

17. code device according to claim 16, it is characterised in that the associated picture block includes and the enhancement layer The adjacent two or more encoded image blocks of image block.

18. the code device according to any one of claim 13 to 17, it is characterised in that the code device also includes Writing unit, for when second determining unit is determined using the vague generalization residual prediction algorithm, by vague generalization residual error In prediction indication position write-in enhancement layer bitstream, wherein, the vague generalization residual prediction flag write in the enhancement layer bitstream Position is set to indicate that the pattern of the vague generalization residual prediction algorithm, determines not using described general in second determining unit When changing residual prediction algorithm, not by vague generalization residual prediction flag position write-in enhancement layer bitstream.

19. a kind of decoding apparatus, it is characterised in that including：

First determining unit, the associated picture block for determining enhancement layer image block, the associated picture block includes enhancement layer figure As the corresponding Primary layer image block of block or the decoded image blocks adjacent with the enhancement layer image block；

Second determining unit, for according to the corresponding residual information of the associated picture block, it is determined that to the enhancement layer image During motion compensation in block perform decoding reconstruction operation, if vague generalization residual prediction algorithm is used, in order to the enhancing Tomographic image block performs the motion compensation to obtain the prediction signal of the enhancement layer image block, and according to the enhancement layer image The prediction signal of block is decoded to the enhancement layer image block, wherein, the corresponding residual information of the associated picture block is used for Indicate whether the associated picture block has residual error or the residual error size degree of the associated picture block.

20. decoding apparatus according to claim 19, it is characterised in that the corresponding residual information bag of the associated picture block Include the coded block flag of the associated picture block, the coded block flag of one or more transform blocks of the associated picture block and institute State at least one of residual error non-zero transform coefficient that one or more transform blocks of associated picture block include.

21. decoding apparatus according to claim 20, it is characterised in that second determining unit specifically for：

22. decoding apparatus according to claim 20, it is characterised in that second determining unit specifically for：

What the one or more transform blocks for including the associated picture block in the corresponding residual information of the associated picture block included During residual error non-zero transform coefficient, and the amplitude of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block It is less than the less than first threshold and the number of the residual error non-zero transform coefficient of one or more transform blocks of the associated picture block When at least one of two threshold values are met, it is determined that the motion in the enhancement layer image block perform decoding reconstruction operation During compensation, the vague generalization residual prediction algorithm is not used.

23. decoding apparatus according to claim 20, it is characterised in that the associated picture block includes and the enhancement layer The two neighboring or more than two decoded image blocks of image block.

24. the decoding apparatus according to any one of claim 19 to 23, it is characterised in that the decoding apparatus also includes Resolution unit, for when second determining unit is determined using the vague generalization residual prediction algorithm, from enhancement layer bitstream Middle parsing vague generalization residual prediction flag position, to determine the vague generalization according to the setting of vague generalization residual prediction flag position The pattern of residual prediction algorithm, second determining unit determine do not use the vague generalization residual prediction algorithm when, not from Vague generalization residual prediction flag position is parsed in enhancement layer bitstream.