CN109644276A - Image coding/decoding method - Google Patents
Image coding/decoding method Download PDFInfo
- Publication number
- CN109644276A CN109644276A CN201780048129.1A CN201780048129A CN109644276A CN 109644276 A CN109644276 A CN 109644276A CN 201780048129 A CN201780048129 A CN 201780048129A CN 109644276 A CN109644276 A CN 109644276A
- Authority
- CN
- China
- Prior art keywords
- block
- current block
- motion vector
- information
- transformation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 112
- 230000033001 locomotion Effects 0.000 claims abstract description 379
- 239000013598 vector Substances 0.000 claims abstract description 288
- 230000009466 transformation Effects 0.000 claims abstract description 220
- 238000013139 quantization Methods 0.000 claims abstract description 74
- 238000006243 chemical reaction Methods 0.000 claims abstract description 25
- 230000002123 temporal effect Effects 0.000 claims description 23
- 230000008859 change Effects 0.000 claims description 19
- 238000005259 measurement Methods 0.000 claims 1
- 238000005070 sampling Methods 0.000 description 35
- 238000010586 diagram Methods 0.000 description 30
- 238000012545 processing Methods 0.000 description 27
- 230000011664 signaling Effects 0.000 description 27
- 238000005192 partition Methods 0.000 description 25
- 239000011159 matrix material Substances 0.000 description 21
- 238000013507 mapping Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 12
- 238000001914 filtration Methods 0.000 description 9
- 238000000638 solvent extraction Methods 0.000 description 9
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 8
- 238000003860 storage Methods 0.000 description 8
- 238000009795 derivation Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 239000000470 constituent Substances 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000008520 organization Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 3
- 230000006835 compression Effects 0.000 description 3
- 238000007906 compression Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011002 quantification Methods 0.000 description 3
- 230000002441 reversible effect Effects 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000036299 sexual function Effects 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- 238000013316 zoning Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/117—Filters, e.g. for pre-processing or post-processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/103—Selection of coding mode or of prediction mode
- H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/119—Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/137—Motion inside a coding unit, e.g. average field, frame or block difference
- H04N19/139—Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
- H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/44—Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/91—Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Physics & Mathematics (AREA)
- Discrete Mathematics (AREA)
- General Physics & Mathematics (AREA)
Abstract
The present invention relates to a kind of methods for realizing motion compensation using motion-vector prediction.A kind of picture decoding method for this method can comprise the following steps that the residual signals for obtaining the quantization for current block；Inverse quantization is carried out to the residual signals of quantization；And it determines for the conversion scheme to residual signals progress inverse transformation.The inverse transformation includes transformation and quadratic transformation for the first time, and at least one of conversion scheme and quadratic transformation scheme can be derived from the decoded reconstructed block that is completed around current block for the first time.
Description
Technical field
The present invention relates to a kind of methods and apparatus for encoding/decoding to image.More particularly it relates to
A kind of encoded information by using contiguous block is come the method and apparatus that derives the encoded information of current block.
Background technique
Recently, for high-resolution and high quality graphic (such as, fine definition (HD) image and ultrahigh resolution (UHD)
Image) demand increase in various application fields.However, the image data of higher resolution and quality and traditional images number
Data volume is increased according to comparing.Therefore, when the medium transmission picture number by using such as traditional wire and wireless broadband network
According to when, or when by using conventional storage media store image data when, the increased costs of transimission and storage.In order to solve with
The resolution ratio and Quality advance and these problems for occurring of image data, need a kind of efficient image coding/decoding technology to use
In higher resolution and higher-quality image.
Image Compression includes various technologies, comprising: from the previous picture of current picture or subsequent picture to being included in
The inter-frame prediction techniques that pixel value in current picture is predicted；By using the Pixel Information in current picture to being included in
The infra-prediction techniques that pixel value in current picture is predicted；The transformation compressed for the energy to residual signals and
Quantification technique；Short code is distributed into the value with the high frequency of occurrences and compiles the entropy that long code distributes to the value with the low frequency of occurrences
Code technology；Etc..Image data can be effectively incompressible by using such Image Compression, and be can be transmitted or deposited
Storage.
In conventional motion compensation, only that spatial motion vectors are candidate, temporal motion vector candidate and zero motion vector are waited
Choosing is added to motion vector candidates list to be used, and single directional prediction and bi-directional predicted is used only, therefore compile to improving
There is limitation in code efficiency.
Summary of the invention
Technical problem
The purpose of the present invention is to provide a kind of to derive the encoded information of current block from the reconstructed block adjacent with current block
Method and apparatus.
Another object of the present invention is to provide a kind of for the motion vector difference and current block adjacent with current block
The method and apparatus that difference between motion vector difference encodes/decodes.
Technical solution
According to the present invention, a kind of image encoding method includes: to generate the prediction signal of current block；It is generated based on prediction signal
The residual signals of current block；Determine the conversion scheme for being converted to residual signals；And quantization is executed to residual signals.
Here, the transformation includes transformation and quadratic transformation, and at least one in conversion scheme and quadratic transformation scheme for the first time for the first time
It is a to be derived from the coding and rebuilding block adjacent with current block.
According to the present invention, a kind of picture decoding method includes: the residual signals for obtaining the quantization of current block；To the residual of quantization
Difference signal executes inverse quantization；And it determines for the conversion scheme to residual signals progress inverse transformation.Here, the inverse transformation packet
Include transformation and quadratic transformation for the first time, and for the first time at least one of conversion scheme and quadratic transformation scheme from adjacent with current block
Decoding and rebuilding block be derived.
In image encoding method or picture decoding method, when prediction signal is generated by intra prediction, become for the first time
Changing at least one of scheme and quadratic transformation scheme can be from intra prediction identical with the intra prediction mode of current block
The contiguous block of mode is derived.
In image encoding method or picture decoding method, when in frame identical with the intra prediction mode of current block
When the conversion scheme for the first time of the contiguous block of prediction mode indicates that transformation is skipped, conversion scheme and the quadratic transformation side for the first time of current block
Case can be confirmed as transformation and skip.
In image encoding method or picture decoding method, quadratic transformation scheme can be from conversion scheme for the first time and current block
The identical contiguous block of conversion scheme is derived for the first time.
In image encoding method or picture decoding method, when prediction signal is generated by inter-prediction, become for the first time
At least one of scheme and quadratic transformation scheme are changed from neighbouring with motion information identical with the motion information of current block
Block is derived.
In image encoding method or picture decoding method, motion information may include motion vector, reference picture indices and
At least one of reference picture direction.
Technical effect
It according to the present invention, can be by providing a kind of encoded information for deriving current block from the reconstructed block adjacent with current block
Method and apparatus improve coding/decoding efficiency.
According to the present invention, can by for the motion vector difference to the motion vector difference and current block adjacent with current block it
Between the difference method and apparatus that encodes/decodes improve coding/decoding efficiency.
Detailed description of the invention
Fig. 1 is the block diagram for showing the configuration of encoding device of embodiment according to the present invention.
Fig. 2 is the block diagram for showing the configuration of decoding device of embodiment according to the present invention.
Fig. 3 is the diagram for schematically showing the partitioned organization of image when coding and decoding to image.
Fig. 4 be show may include predicting unit (PU) in coding unit (CU) form diagram.
Fig. 5 be show may include converter unit (TU) in coding unit (CU) form diagram.
Fig. 6 is the diagram for the embodiment for explaining the processing of intra prediction.
Fig. 7 is the diagram for the embodiment for explaining the processing of inter-prediction.
Fig. 8 is the diagram for explaining the mapping ensemble according to intra prediction mode.
Fig. 9 is the diagram for explaining the processing of transformation.
Figure 10 is the diagram for explaining the scanning to the transformation coefficient by quantization.
Figure 11 is the diagram for explaining block subregion.
Figure 12 is the exemplary diagram for showing the coding/decoding unit according to the zoned format of block.
Figure 13 is the flow chart for the processing for depicting a determination whether that the information to binary tree subregion is decoded.
Figure 14 is the flow chart for the processing for depicting a determination whether that the information to binary tree subregion is decoded.
Figure 15 to Figure 17 is shown no longer for the feelings with predetermined size or smaller size of piece of execution binary tree subregion
The exemplary diagram of condition.
Figure 18 is to show to determine whether to derive the residual of current block from contiguous block when current block is encoded by intra prediction
The flow chart of the processing of the encoded information of difference signal.
Figure 19 is to show to determine whether to derive the residual of current block from contiguous block when current block is encoded by inter-prediction
The flow chart of the processing of the encoded information of difference signal.
Figure 20 is the flow chart for showing the decoding process of motion vector of current block.
Figure 21 is to show the exemplary diagram for deriving spatial motion vectors candidate.
Figure 22 is to show the exemplary diagram for deriving temporal motion vector candidate.
Figure 23 is to show the diagram for deriving the second motion vector difference.
Specific embodiment
Invention mode
A variety of modifications can be made to the present invention, and there are various embodiments of the invention, wherein now with reference to attached drawing
The examples of the embodiments are provided and will be described in the examples of the embodiments.However, the invention is not limited thereto, although showing
Example property embodiment can be interpreted as including all modifications, equivalent form or replacement in technical concept and technical scope of the invention
Form.Similar reference label refers to the same or similar function in all respects.In the accompanying drawings, for the sake of clarity, the shape of element
Shape and size can be exaggerated.In the following detailed description of the invention, to shown by way of diagram can to the present invention carry out
The attached drawing of the specific embodiment of implementation carries out reference.These embodiments are described in enough detail so that those skilled in the art's energy
Enough implement the disclosure.It should be understood that the various embodiments of the disclosure need not be mutually exclusive although difference.For example, not
In the case where being detached from spirit and scope of the present disclosure, described herein and the associated special characteristic of one embodiment, structure and spy
Property can be carried out in other embodiments.Moreover, it should be understood that without departing from the spirit and scope of the disclosure, often
The position of each element in a disclosed embodiment or arrangement can be modified.Therefore, described in detail below will be not with limitation
Meaning carries out, and the scope of the present disclosure only (in the case where suitable explain, is also wanted together with claim by appended claims
Seek the full scope of the equivalent of protection) it limits.
Term " first " used in the description, " second " etc. can be used for describing various assemblies, but these components are simultaneously
It is not construed as limited to the term.The term is only applied to distinguish a component with another component.For example, not
In the case where departing from the scope of the present invention, " first " component is referred to alternatively as " second " component, and " second " component can also be by class
As be known as " first " component.Term "and/or" include multiple combination or multiple items in any one.
It will be appreciated that in the present specification, be referred to simply as " being connected to " or " being integrated to " another element when element and
When not being " being directly connected to " or " being bonded directly to " another element, it " can be directly connected to " or " being bonded directly to " is another
Element, or another element is connected to or is integrated in the case where being inserted into other elements therebetween.On the contrary, it will be appreciated that when
When element referred to as " is bound directly " or " being directly connected to " arrives another element, intermediary element is not present.
In addition, the building block shown in an embodiment of the present invention is shown separately, so that spy different from each other is presented
Sexual function.Therefore, this is not meant to that each building block is configured with the component units of individual hardware or software.In other words
It says, for convenience, each building block includes each of the building block enumerated.Therefore, in each building block extremely
Few two building blocks, which can be combined to form a building block or a building block, can be divided into multiple building blocks
To execute each function.It is no be detached from essence of the invention in the case where, embodiment that each building block is combined and
One divided embodiment of building block is also included in the scope of the present invention.
The term used in the present specification is only used for description specific embodiment, is not intended to limit the invention.With odd number
The expression used includes plural number expression, unless it has visibly different meaning within a context.In the present specification, it will manage
Solution, the term of " including ... ", " having ... " etc. be intended to indicate feature disclosed in the description, quantity, step,
Behavior, component, assembly unit, or combinations thereof presence, and be not intended to exclude one or more other features, quantity, step, row
For, component, assembly unit, or combinations thereof there may be or a possibility that may be added.In other words, when particular element is referred to as
When " by including ", the element in addition to respective element is not excluded, but, element in addition can be included in reality of the invention
It applies in example or the scope of the present invention.
In addition, some constituent element may not be the indispensable constituent element for executing necessary function of the invention, but
Only promote the optional constituent element of its performance.It can be by only including substantive indispensable building block for carrying out the present invention
And the building block used in improving performance is excluded to implement the present invention.Only include the indispensable building block and excludes
The structure of the optional building block used in only improving performance is also included in the scope of the present invention.
Hereinafter, it will be described in detail with reference to the accompanying drawings the embodiment of the present invention.In description exemplary embodiment of the present invention
When, known function or structure will not be discussed in detail, this is because they can unnecessarily be obscured the present invention.In attached drawing
Identical constituent element indicated by identical reference label, and the repeated description of similar elements will be omitted.
In addition, hereinafter, image can mean the picture for constituting video, or can mean video itself.For example, " to figure
As being encoded or decoded or coded and decoded " can mean " video is encoded or is decoded or is carried out coding and
Decoding ", and can mean and " be encoded or decoded or carried out coding reconciliation to an image among the multiple images of video
Code ".Here, picture and image can have the same meaning.
Term description
Encoder: the equipment for executing coding can be meant.
Decoder: it can mean and execute decoded equipment.
Parsing: it can mean by executing entropy decoding and determine the value of syntactic element, or can mean entropy decoding itself.
Block: M × N matrix sampling point can be meant.Here, M and N is positive integer, and block can mean the sampling point of two dimensional form
Matrix.
Sampling point: being the basic unit of block, and can indicate according to bit-depth (Bd) and value that range is 0 to 2Bd -1.Sample
Point can mean pixel in the present invention.
Unit: the unit coded and decoded to image can be meant.When coding and decoding to image, unit can
To be the region generated and carrying out subregion to an image.In addition, unit can be meant during coding or decoding when one
Image is partitioned sub- division unit when multiple sub- division units.It is executable to be directed to when being coded and decoded to image
The predetermined process of each unit.One unit can be partitioned smaller subelement of the size than the unit.Foundation function,
Unit can mean that block, macro block, coding tree unit, coding tree block, coding unit, encoding block, predicting unit, prediction block, transformation are single
Member, transform block etc..In addition, unit may include the coloration point of luminance component block, luminance component block in order to distinguish unit and block
The syntactic element of gauge block and each color component block.Unit can have various sizes and shape, specifically, the shape of unit
Shape can be two-dimentional geometric figure, rectangle, square, trapezoidal, triangle, pentagon etc..In addition, unit information may include
Cell type (instruction coding unit, predicting unit, converter unit etc.), unit size, unit depth, unit is carried out coding and
At least one of decoded sequence etc..
It rebuilds adjacent unit: can mean and be formerly encoded on spatially/time or decoded reconstruction unit, and rebuild
Unit is adjacent with coding/decoding object element.Here, it is gratifying for reconstruction contiguous block to rebuild adjacent unit.
Contiguous block: the block adjacent with coding/decoding object block can be meant.The block adjacent with coding/decoding object block is gratifying
For the block with the boundary contacted with coding/decoding object block.Contiguous block can mean the adjacent top positioned at coding/decoding object block
The block of point.Contiguous block is gratifying for reconstruction contiguous block.
Unit depth: can mean unit is partitioned degree.In tree construction, root node can be highest node, leaf segment
Point can be minimum node.
Symbol: syntactic element, coding parameter, the value of transformation coefficient of coding/decoding object element etc. can be meant.
Parameter set: the head information in the structure of bit stream can be meant.Parameter set may include video parameter collection, sequential parameter
At least one parameter set in collection, parameter sets or auto-adaptive parameter set.In addition, parameter set can mean slice header information and simultaneously
Row block (tile) head information etc..
Bit stream: the Bit String including coded image information can be meant.
Predicting unit: it can mean substantially single when executing inter prediction or infra-frame prediction and for the compensation of prediction
Member.One predicting unit can be partitioned multiple subregions.In this case, each subregion in multiple subregions, which can be, is holding
Basic unit when row prediction and compensation, and can be predicting unit from each subregion that predicting unit subregion goes out.In addition, one
A predicting unit can be partitioned multiple small predicting units.Predicting unit can have various sizes and shape, and specifically come
It says, the shape of predicting unit can be two-dimentional geometric figure, rectangle, square, trapezoidal, triangle, pentagon etc..
Predicting unit subregion: the shape for the predicting unit that subregion goes out can be meant.
Reference picture list: can mean the list including at least one reference picture, wherein it is described at least one refer to picture
Face is used for inter-prediction or motion compensation.The type of reference picture list can be List Combined (LC), List 0
(L0), List 1 (L1), List 2 (L2), List 3 (L3) etc..At least one reference picture list can be used for inter-prediction.
Inter-prediction indicator: gratifying to be one of the following: the interframe of the coding/decoding object block in the case of inter-prediction is pre-
Survey direction (single directional prediction, bi-directional predicted etc.), for the reference picture by coding/decoding object block generation prediction block
The quantity of quantity and the reference block for executing inter-prediction or motion compensation by the coding/decoding object block.
Reference picture indices: the index with particular reference to picture in reference picture list can be meant.
Reference picture: discrete cell can be meant for picture referenced by inter-prediction or motion compensation.Reference picture can
Referred to as reference picture.
Motion vector: being the two-dimensional vector for inter-prediction or motion compensation, and can mean that coding/decoding target is drawn
Offset between face and reference picture.For example, (mvX, mvY) can indicate that motion vector, mvX can indicate horizontal component, mvY be can refer to
Show vertical component.
Motion vector candidates: it can mean the unit when predicted motion vector as predicting candidate, or can mean the list
The motion vector of member.
Motion vector candidates list: the list configured by using motion vector candidates can be meant.
Motion vector candidates index: the indicator of the motion vector candidates in instruction motion vector candidates list can be meant.
Motion vector candidates index the index for being referred to alternatively as motion vector predictor.
Motion information: motion vector, reference picture indices and inter-prediction indicator can be meant and including reference picture
The information of at least one of list information, reference picture, motion vector candidates, motion vector candidates index etc..
Merge candidate list: can mean by using the list for merging candidate configuration.
Merge candidate: may include that space merging is candidate, time merging is candidate, combination merging is candidate, combined bidirectional is predicted to close
And candidate, zero merging candidate etc..Merging candidate may include the motion information of such as prediction type, for the ginseng of each list
Examine picture index, motion vector etc..
Merge index: can mean that instruction merges the candidate information of the merging in candidate list.In addition, merging index can indicate
Derivation among the reconstructed block adjacent in space time with current block merges candidate block.In addition, merging index can indicate to close
And at least one motion information in candidate a plurality of motion information.
Converter unit: it can mean and be executed and transformation, inverse transformation, quantization, inverse quantization and transformation coefficient when to residual signals
Basic unit when the similar coding/decoding of coding/decoding.One converter unit can be partitioned multiple small converter units.Become
Various sizes and shape can be had by changing unit.Specifically, the shape of converter unit can be two-dimentional geometric figure, such as square
Shape, square, trapezoidal, triangle, pentagon etc..
Scaling: the processing that Graph One factor is multiplied with transform coefficient levels can be meant, as a result, transformation coefficient can be produced
It is raw.Scaling may be additionally referred to as inverse quantization.
Quantization parameter: the value used during quantifying with inverse quantization when zooming in and out to transform coefficient levels can be meant.
Here, quantization parameter can be mapped to the value of the step sizes of quantization.
Increment (Delta) quantization parameter: it can mean that the quantization parameter of coding/decoding object element and the quantization predicted are joined
Difference between number.
Scanning: the method being ranked up to the coefficient order in block or matrix can be meant.For example, two-dimensional matrix is ordered as
The operation of one-dimensional matrix is referred to alternatively as scanning, and by one-dimensional matrix sort be two-dimensional matrix operation be referred to alternatively as scanning or it is inverse
Scanning.
Transformation coefficient: the coefficient value generated after executing transformation can be meant.In the present invention, by the transformation series of quantization
Number grade (being applied the transformation coefficient of quantization) is referred to alternatively as transformation coefficient.
Non-zero transform coefficient: can mean value not be 0 transformation coefficient, or can mean value not for 0 transform coefficient levels.
Quantization matrix: it can mean in quantization and inverse quantization using the main quality (subject to improve image
) or the matrix of object quality (object quality) quality.Quantization matrix is referred to alternatively as scaling list.
Quantization matrix coefficient: each element of quantization matrix can be meant.Quantization matrix coefficient is referred to alternatively as matrix coefficient.
Default matrix: the predetermined quantitative matrix being predefined in the encoder and the decoder can be meant.
Non-default matrix: it can mean and be sent/connect by user in the case where not being predefined in the encoder and the decoder
The quantization matrix of receipts.
Coding tree unit: can be by luminance component (Y) coding tree unit and relevant two chromatic components (Cb, Cr)
Coding tree unit is constituted.Each coding tree unit can be by using at least one partition method (quaternary tree, binary tree etc.)
It is partitioned, to constitute subelement, coding unit, predicting unit, converter unit etc..Coding tree unit is used as referring to
Show the art of block of pixels (wherein, block of pixels is the processing unit in the decoding/encoding processing of image, such as the subregion of input picture)
Language.
Coding tree block: it can be used as the art for being used to indicate one of Y coding tree unit, Cb coding tree unit and Cr coding tree unit
Language.
Fig. 1 is the block diagram for showing the configuration of encoding device of embodiment according to the present invention.
Encoding device 100 can be video encoder or image encoding apparatus.Video may include one or more figures
Picture.Encoding device 100 can sequentially in time encode one or more images of video.
Referring to Fig.1, encoding device 100 may include motion prediction unit 111, motion compensation units 112, intraprediction unit
120, switch 115, subtracter 125, converter unit 130, quantifying unit 140, entropy code unit 150, inverse quantization unit 160,
Inverse transformation block 170, adder 175, filter cell 180 and reference picture buffer 190.
Encoding device 100 can be according to frame mode or inter-frame mode or both frame mode and inter-frame mode come to input
Picture is encoded.In addition, encoding device 100 can generate bit stream, and exportable production by being encoded to input picture
Raw bit stream.When frame mode is used as prediction mode, switch 115 be can switch in frame.When inter-frame mode is used as
When prediction mode, switch 115 can switch to interframe.Here, frame mode is referred to alternatively as intra prediction mode, and inter-frame mode can
Referred to as inter-frame forecast mode.Encoding device 100 can produce the prediction block of the input block of input picture.In addition, generating prediction
After block, encoding device 100 can encode the residual error between input block and prediction block.Input picture is referred to alternatively as working as
The present image of the target of preceding coding.Input block is referred to alternatively as current block or is referred to alternatively as the volume of the target as present encoding
Code object block.
When prediction mode is frame mode, intraprediction unit 120 can be by the previous coding block adjacent with current block
Pixel value is used as reference pixel.Intraprediction unit 120 can usually execute spatial prediction by using reference image, and can be by making
The prediction sampling point of input block is generated with spatial prediction.Here, intra prediction can mean that intra frame is predicted.
When prediction mode is inter-frame mode, motion prediction unit 111 can be searched in motion prediction process from reference picture
The region of rope and input block Optimum Matching, and motion vector can be derived by using the region searched.Reference picture can be deposited
Storage is in reference picture buffer 190.
Motion compensation units 112 can execute motion compensation by using motion vector to generate prediction block.Here, movement arrow
Amount can be the two-dimensional vector for inter-prediction.In addition, motion vector can indicate it is inclined between current picture and reference picture
It moves.Here, inter-prediction can mean that inter-frame is predicted.
When the value of motion vector is not integer, motion prediction unit 111 and motion compensation units 112 can be by references
Partial region in picture generates prediction block using interpolation filter.In order to execute inter-prediction or movement based on coding unit
Compensation can determine the prediction in coding unit among skip mode, merging patterns, AMVP mode and current picture reference model
Which kind of method the motion prediction of unit and compensation method use.Inter-prediction or motion compensation can be executed according to each pattern.This
In, current picture reference model can mean the prediction mould in the region rebuild in advance using the current picture with encoding target block
Formula.In order to indicate the region rebuild in advance, the motion vector for current picture reference model can define.Whether according to working as
Preceding picture reference model is encoded encoding target block and can be encoded by using the reference picture indices of encoding target block.
Subtracter 125 can generate residual block by using the residual error between input block and prediction block.Residual block can be claimed
For residual signals.
Converter unit 130 can generate transformation coefficient, and exportable transformation coefficient by being converted to residual block.This
In, transformation coefficient can be the coefficient value generated and converting to residual block.In the case where converting skip mode, converter unit
130 can skip the transformation to residual block.
It can be by generating the transform coefficient levels by quantization to transformation coefficient application quantization.Hereinafter, in this hair
In bright embodiment, transformation coefficient is referred to alternatively as by the transform coefficient levels of quantization.
Quantifying unit 140 can generate the transformation series by quantization by being quantified according to quantization parameter to transformation coefficient
Number grade, and the exportable transform coefficient levels by quantization.Here, quantifying unit 140 can be come pair by using quantization matrix
Transformation coefficient is quantified.
Entropy code unit 150 can be by being worth or to by quantifying unit 140 is calculated in coding according to probability distribution
Calculated encoded parameter values etc. execute entropy coding to generate bit stream, and the bit stream of exportable generation in reason.Entropy coding list
Member 150 can execute entropy coding to the information for being decoded to image, and execute entropy coding to the information of the pixel of image.Example
Such as, for may include syntactic element etc. to the information that image is decoded.
When entropy coding is by application, by distributing a small amount of bit to the symbol with high probability of occurrence and to low appearance
The symbol of probability distributes a large amount of bits to indicate symbol, to reduce the size of the bit stream encoded to aiming symbol.Cause
This, by entropy coding, the compression performance of image coding be can be improved.For entropy coding, such as index is can be used in entropy code unit 150
The coding of Columbus, context-adaptive variable-length encoding (CAVLC) and context adaptive binary arithmetic coding (CABAC)
Method.For example, entropy code unit 150 can execute entropy coding by using variable-length encoding/code (VLC) table.In addition, entropy coding list
Member 150 can derived object symbol binarization method and aiming symbol/binary digit probabilistic model, and can then pass through
Arithmetic coding is executed using the binarization method derived or the probabilistic model derived.
In order to encode to transform coefficient levels, entropy code unit 150 can be incited somebody to action by using transformation coefficient scan method
The coefficient of two-dimensional block form changes into one-dimensional vector form.For example, by scanning the coefficient come scanning block, two dimensional form with upper right
Coefficient can be changed to a n dimensional vector n.According to the size and intra prediction mode of converter unit, can be used for along column side
Vertical scan direction to the coefficient of scanning two-dimensional block form and the water for the coefficient along line direction scanning two-dimensional block form
Square to scanning, rather than scanned using upper right.That is, size and intra prediction mode according to converter unit, it can
Determine which kind of scan method among upper right scanning, vertical scan direction and horizontal direction scanning will be used.
Coding parameter may include the information for encoding and being sent to such as syntactic element of decoder by encoder, and can wrap
Include the information that can be derived in coding or decoding process.Necessary to coding parameter can mean and be encoded or be decoded to image
Information.For example, coding parameter may include at least one of following item value or combining form: block size, block depth, block subregion letter
Breath, unit size, unit depth, unit partition information, the subregion sign of quaternary tree form, binary tree form subregion sign,
The subregion direction of binary tree form, intra prediction mode, intra prediction direction, with reference to sampling point filtering method, prediction block boundary filter
Wave method, filter tap, filter coefficient, inter-frame forecast mode, motion information, motion vector, reference picture indices, interframe
Prediction direction, inter-prediction indicator, reference picture list, motion vector predictor, motion vector candidates list, about fortune
Whether information that whether dynamic merging patterns are used, movement merge candidate, movement and merge candidate list, made about skip mode
Accuracy that information, interpolation filter type, motion vector size, motion vector indicate, alternative types, transform size,
The information whether used about additional (secondary) transformation, the information, the coded block pattern, coding that whether there is about residual signals
Block mark, quantization parameter, quantization matrix, the filter information in loop, the letter whether applied in loop about filter
Breath, the filter coefficient in loop, binaryzation/anti-binarization method, context model, context binary digit, bypass two into
Position processed, transformation coefficient, transform coefficient levels, transform coefficient levels scan method, image show/output sequence, band identification letter
Breath, type of strip, band partition information, parallel block identification information, parallel block type, parallel block partition information, picture type, ratio
The information of special depth and luminance signal or carrier chrominance signal.
Residual signals can mean the difference between original signal and prediction signal.Selectively, residual signals, which can be, passes through
The signal that difference between original signal and prediction signal is converted and is generated.Selectively, residual signals, which can be, passes through
The signal that transform and quantization is carried out to the difference between original signal and prediction signal and is generated.Residual block can be the residual of module unit
Difference signal.
When encoding device 100 is executed by using inter-prediction to be encoded.The current picture of coding is used as being directed to and incite somebody to action
The reference picture for another image being then processed.Therefore, encoding device 100 can be decoded the current picture of coding, and
Decoded image can be stored as reference picture.In order to execute decoding, inverse quantization and inversion can be executed to the current picture of coding
It changes.
The coefficient of quantization can be by inverse quantization unit 160 by inverse quantization, and can be inversely transformed by inverse transformation block 170.
It can will be added by inverse quantization and the coefficient of inverse transformation with prediction block by adder 175, thus can produce reconstructed block.
Reconstructed block can pass through filter cell 180.Filter cell 180 can be to reconstructed block or reconstruction picture application deblocking filter
At least one of wave device, sampling point self adaptation skew (SAO) (SAO) and auto-adaptive loop filter (ALF).Filter cell 180 can
Referred to as loop filter.
De-blocking filter can remove the block distortion that boundary between blocks occurs.In order to determine de-blocking filter whether by
Operation can determine whether de-blocking filter is applied to current block based on including the pixel in several row or column in block.
When de-blocking filter is applied to block, strong filter or weak filter can be applied according to required deblocking filtering intensity.This
It outside, can the filtering of parallel processing horizontal direction and vertical direction filtering in application de-blocking filter.
Optimal deviant can be added to pixel value to compensate to encoding error by sampling point self adaptation skew (SAO).Sampling point is certainly
The offset between the image and raw frames Jing Guo deblocking filtering can be corrected for each pixel by adapting to offset.In order to right
Specific picture executes offset correction, the method that the marginal information for considering each pixel can be used to apply offset, or uses following
Method: being the region of predetermined quantity by the pixel partitions of image, the region of offset correction will be performed by determining, and to identified
Apply offset correction in region.
Auto-adaptive loop filter can be held based on the value obtained will be compared with raw frames by will rebuild picture
Row filtering.The pixel of image can be partitioned predetermined group, be applied to each group of a filter and be determined, and is different
Filtering can be performed at each group.The information for whether being applied to luminance signal about auto-adaptive loop filter can be for each
Coding unit (CU) is sent.Shape and the filter coefficient for being applied to each piece of auto-adaptive loop filter are alterable.
In addition, with same form (fixed form) auto-adaptive loop filter can in the case where not considering the characteristic of object block quilt
Using.
It can be stored in reference picture buffer 190 by the reconstructed block of filter cell 180.
Fig. 2 is the block diagram for showing the configuration of decoding device of embodiment according to the present invention.
Decoding device 200 can be video decoding apparatus or image decoding apparatus.
Referring to Fig. 2, decoding device 200 may include entropy decoding unit 210, inverse quantization unit 220, inverse transformation block 230, frame
Interior prediction unit 240, motion compensation units 250, adder 255, filter cell 260 and reference picture buffer 270.
Decoding device 200 can receive the bit stream exported from encoding device 100.Decoding device 200 can be according to frame mode
Or inter-frame mode decodes bit stream.In addition, decoding device 100 can generate reconstruction picture by executing decoding, and can
Picture is rebuild in output.
When the prediction mode used in decoding is frame mode, switch can be switched in frame.When in decoding
When the prediction mode used is inter-frame mode, switch can be switched to interframe.
Decoding device 200 can obtain reconstructive residual error block from the bit stream of input, and can produce prediction block.When reconstructive residual error block
When being obtained with prediction block, decoding device 200 can be by generating reconstructive residual error block and prediction block phase Calais as decoding target
The reconstructed block of block.Decoding object block is referred to alternatively as current block.
Entropy decoding unit 210 can generate symbol by executing entropy decoding to bit stream according to probability distribution.The symbol of generation
It number may include the symbol of the transform coefficient levels with quantization.Here, the method for entropy decoding can be with the method class of above-mentioned entropy coding
Seemingly.For example, the method for entropy decoding can be the inversely processing of the method for above-mentioned entropy coding.
In order to be decoded to transform coefficient levels, transformation coefficient scanning is can be performed in entropy decoding unit 210, one-dimensional as a result,
The coefficient of vector form can be changed to two-dimensional block form.For example, by scanning the coefficient come scanning block, a n dimensional vector n with upper right
The coefficient of form can be changed to two-dimensional block form.According to the size and intra prediction mode of converter unit, can be used vertical
Direction scanning and horizontal direction scanning, rather than scanned using upper right.That is, size and frame according to converter unit
Inner estimation mode, it may be determined which kind of scan method among upper right scanning, vertical scan direction and horizontal direction scanning is made
With.
Transform coefficient levels by quantization can be by inverse quantization unit 220 by inverse quantization, and can pass through inverse transformation block
230 are inversely transformed.Transform coefficient levels by quantization are by inverse quantization and are inversely transformed to generate reconstructive residual error block.Here,
Inverse quantization unit 220 can be to the transform coefficient levels application quantization matrix by quantization.
When frame mode quilt in use, intraprediction unit 240 can generate prediction block by executing spatial prediction,
In, spatial prediction uses the pixel value of the early decoding block adjacent with decoding object block.
When inter-frame mode quilt in use, motion compensation units 250 can generate prediction block by executing motion compensation,
In, motion compensation uses both reference picture and the motion vector being stored in reference picture buffer 270.Work as motion vector
Value when not being integer, motion compensation units 250 can be by producing the partial region in reference picture using interpolation filter
Raw prediction block.It, can be in skip mode, merging patterns, AMVP mode and current picture based on coding unit in order to execute motion compensation
Determine which kind of method the motion compensation process of the predicting unit in coding unit uses among the reference model of face.In addition, can foundation
The mode executes motion compensation.Here, current picture reference model can be meant using the current picture with decoding object block
The prediction mode of interior previous reconstruction regions.Previous reconstruction regions can not be adjacent with decoding object block.In order to indicate previous reconstruction
Region can use fixed vector for current picture reference model.In addition, whether instruction decoding object block is according to current picture
The mark or index that decoded piece of reference model can be sent with signal, and can be by using the reference picture of decoding object block
It indexes and is derived.It may be present in the reference picture for decoding object block for the current picture of current picture reference model
Fixation position (for example, position or last position that reference picture indices are 0) in list.In addition, current picture can
Become ground to be located in reference picture list, for this purpose, available signal sends the reference picture indices of the position of instruction current picture.
Reconstructive residual error block can be added with prediction block by adder 255.By the way that reconstructive residual error block is added with prediction block
And the block generated can pass through filter cell 260.Filter cell 260 can be to reconstructed block or reconstruction picture application deblocking filtering
At least one of device, sampling point self adaptation skew (SAO) and auto-adaptive loop filter.The exportable reconstruction picture of filter cell 260.
Rebuilding picture can be stored in reference picture buffer 270, and can be used for inter-prediction.
Fig. 3 is the diagram for schematically showing the partitioned organization of the image when coding and decoding to image.Fig. 3 shows
Embodiment by a unit subregion for multiple subelements is shown to meaning property.
Efficient zoned in order to carry out to image, coding unit (CU) can be used in coding and decoding.Here, coding is single
Member can mean the unit encoded.Unit can be 1) syntactic element and 2) include image sampling point block combination.For example,
" subregion of unit " can mean " relevant to unit piece of subregion ".Block partition information may include the information about unit depth.
Depth information can the degree or both that is partitioned of the number that is partitioned of indicating unit or unit.
Referring to Fig. 3, image 300 is directed to each maximum coding unit (LCU) by sequential partition, and partitioned organization is for every
A LCU is determined.Here, LCU and coding tree unit (CTU) have the same meaning.One unit can have based on tree construction
Depth information, and can be by layering and zoning.The subelement that each subregion goes out can have depth information.Depth information indicating unit quilt
The degree or both that the number or unit of subregion are partitioned, therefore, depth information may include the information of the size about subelement.
Partitioned organization can mean the distribution of the coding unit (CU) in LCU 310.CU can be for having to image
Imitate the unit of coding.The distribution can based on CU whether will by repeatedly (positive integer equal to or more than 2, including 2,4,8,
16 etc.) subregion and be determined.The width dimensions and height dimension of the CU that subregion goes out may respectively be the half width dimensions of original CU
With half height dimension.Selectively, the width dimensions and height dimension of the CU gone out according to the number of subregion, subregion can be small respectively
In the width dimensions and height dimension of original CU.The CU that subregion goes out can be the CU that multiple further subregions go out by recursively subregion,
Wherein, according to identical partition method, the CU that the further subregion goes out have the CU gone out than the subregion width dimensions and
The smaller width dimensions of height dimension and height dimension.
Here, the subregion of CU can be executed recursively until predetermined depth.Depth information can be the size of instruction CU
Information, and can be stored in each CU.For example, the depth of LCU can be 0, and the depth of minimum coding unit (SCU)
It can be predetermined depth capacity.Here, LCU, which can be, has maximum sized coding unit as described above, and SCU can be with
It is the coding unit with minimum dimension.
When LCU 310 starts to be partitioned, and the width dimensions of CU and height dimension are reduced by division operation,
The depth of CU increases by 1.In the case where the CU that cannot be partitioned, CU can have 2N × 2N size for each depth.Can
In the case where the CU being partitioned, the CU with 2N × 2N size can be partitioned the CU of multiple N × N sizes.Whenever depth increases by 1
When, N's is half-sized.
For example, when a coding unit is partitioned four sub- coding units, one of described four sub- coding units
Width dimensions and height dimension may respectively be the half width dimensions and half height dimension of original coding unit.For example, when 32
When the coding unit of × 32 sizes is partitioned four sub- coding units, each of described four sub- coding units can have
16 × 16 sizes.When a coding unit is partitioned four sub- coding units, coding unit can be according to quaternary tree form quilt
Subregion.
For example, when a coding unit is partitioned two sub- coding units, one of described two sub- coding units
Width dimensions or height dimension may respectively be the half width dimensions or half height dimension of original coding unit.For example, when 32
The coding unit of × 32 sizes by vertical partitioning be two sub- coding units when, each of described two sub- coding units can
With 16 × 32 sizes.For example, when the coding unit of 32 × 32 sizes is two sub- coding units by horizontal partitioning, described two
Each of a sub- coding unit can have 32 × 16 sizes.When a coding unit is partitioned two sub- coding units
When, coding unit can be partitioned according to binary tree form.
Referring to Fig. 3, the size of the LCU with minimum-depth 0 can be 64 × 64 pixels, and with depth capacity 3
The size of SCU can be 8 × 8 pixels.Here, the CU (that is, LCU) with 64 × 64 pixels can be indicated by depth 0, it can be by depth
Degree 1 indicates the CU with 32 × 32 pixels, can indicate CU with 16 × 16 pixels by depth 2, and can by depth 3
Indicate the CU (that is, SCU) with 8 × 8 pixels.
In addition, the information that whether will be partitioned about CU can be indicated by the partition information of CU.Partition information can be 1
Bit information.Partition information can be included in all CU in addition to SCU.For example, CU can when the value of partition information is 0
It is not partitioned, when the value of partition information is 1, CU can be partitioned.
Fig. 4 is the diagram for showing the form for the predicting unit (PU) that can be included in coding unit (CU).
At least one predicting unit can be partitioned from the CU being no longer partitioned among multiple CU that LCU subregion goes out
(PU).The processing is also referred to as subregion.
PU can be the basic unit for prediction.PU can be according to appointing in skip mode, inter-frame mode and frame mode
One mode and be encoded and decoding.PU can be partitioned according to mode according to various forms.
In addition, coding unit can not be partitioned multiple predicting units, and coding unit and predicting unit are with identical
Size.
As shown in figure 4, CU can not be partitioned in skip mode.In skip mode, it can support and not sectorized CU
2N × 2N mode 410 with identical size.
In inter-frame mode, 8 compartment models can be supported in CU.For example, can support 2N × 2N mould in inter-frame mode
Formula 410,2N × N mode 415, N × 2N mode 420, N × N mode 425,2N × nU mode 430,2N × nD mode 435, nL ×
2N mode 440 and nR × 2N mode 445.In frame mode, 2N × 2N mode 410 and N × N mode 425 can be supported.
One coding unit can be partitioned one or more predicting units.One predicting unit can be partitioned one
Or more sub- predicting unit.
For example, when a predicting unit is partitioned four sub- predicting units, one of described four sub- predicting units
Width dimensions and height dimension can be the half width dimensions and half height dimension of original predictive unit.For example, when 32 × 32
When the predicting unit of size is partitioned four sub- predicting units, each of described four sub- predicting units can have 16 ×
16 sizes.When a predicting unit is partitioned four sub- predicting units, predicting unit can be partitioned according to quaternary tree form.
For example, when a predicting unit is partitioned two sub- predicting units, one of described two sub- predicting units
Width dimensions or height dimension can be the half width dimensions or half height dimension of original predictive unit.For example, when 32 × 32
The predicting unit of size by vertical partitioning be two sub- predicting units when, each of described two sub- predicting units can have
16 × 32 sizes.For example, when the predicting unit of 32 × 32 sizes is two sub- predicting units by horizontal partitioning, described two sons
Each of predicting unit can have 32 × 16 sizes.When a predicting unit is partitioned two sub- predicting units, in advance
Surveying unit can be partitioned according to binary tree form.
Fig. 5 is the diagram for showing the form for the converter unit (TU) that can be included in coding unit (CU).
Converter unit (TU) can be in CU for convert, quantify, the basic unit of inverse transformation and inverse quantization.TU can have
There are square shape or rectangular shape etc..TU can be determined independently according to the size of CU or the form of CU or both of the above.
At least one TU can be partitioned from the CU being no longer partitioned among the CU that LCU subregion goes out.Here, the subregion of TU
Structure can be quad-tree structure.For example, as shown in figure 5, a CU 510 can be partitioned according to quad-tree structure it is primary or
More times.The case where one CU is partitioned at least once is referred to alternatively as recursive partitioning.By carrying out subregion, CU 510 can be by
TU of various sizes is formed.Selectively, CU can according to CU carry out subregion vertical line quantity or CU is divided
The horizontal quantity in area or both and be partitioned at least one TU.CU can be partitioned the TU being mutually symmetrical, or can quilt
Subregion is TU asymmetrical relative to one another.In order to be the TU being mutually symmetrical by CU subregion, the information of the size/shape of TU can be by with signal
It sends, and can be derived from the information of the size/shape of CU.
In addition, coding unit can not be partitioned converter unit, and coding unit and converter unit can be having the same
Size.
One coding unit can be partitioned at least one converter unit, and a converter unit can be partitioned at least
One sub- converter unit.
For example, when a converter unit is partitioned four sub- converter units, one of described four sub- converter units
Width dimensions and height dimension may respectively be the half width dimensions and half height dimension of original transform unit.For example, when 32
When the converter unit of × 32 sizes is partitioned four sub- converter units, each of described four sub- converter units can have
16 × 16 sizes.When a converter unit is partitioned four sub- converter units, converter unit can be according to quaternary tree form quilt
Subregion.
For example, when a converter unit is partitioned two sub- converter units, one of described two sub- converter units
Width dimensions or height dimension may respectively be the half width dimensions or half height dimension of original transform unit.For example, when 32
The converter unit of × 32 sizes by vertical partitioning be two sub- converter units when, each of described two sub- converter units can
With 16 × 32 sizes.For example, when the converter unit of 32 × 32 sizes is two sub- converter units by horizontal partitioning, described two
Each of a sub- converter unit can have 32 × 16 sizes.When a converter unit is partitioned two sub- converter units
When, converter unit can be partitioned according to binary tree form.
When executing transformation, residual block can be become by using at least one of predetermined map method transform method
It changes.For example, the predetermined map method may include discrete cosine transform (DCT), discrete sine transform (DST), KLT etc..It can lead to
It crosses using at least one of following item and determines which kind of transform method is applied to convert residual block: predicting unit
The size/shape of inter-frame forecast mode information, the intraprediction mode information of predicting unit and transform block.Indicate transform method
Information can be sent with signal.
Fig. 6 is the diagram for the embodiment for explaining the processing of intra prediction.
Intra prediction mode can be with right and wrong direction mode or direction mode.Non- direction mode can be DC mode or plane mould
Formula.Direction mode can be the prediction mode with specific direction or angle, and the quantity of direction mode can be equal to or
M greater than 1.Direction mode can be indicated as at least one of MODE NUMBER, mode value and pattern angles.
The quantity of intra prediction mode can be the N equal to or more than 1, including non-direction mode and direction mode.
The quantity of intra prediction mode can according to block size and change.For example, when the size of block is 4 × 4 or 8 × 8,
The quantity of intra prediction mode can be 67, and when the size of block is 16 × 16, the quantity of intra prediction mode can be 35, when
When the size of block is 32 × 32, the quantity of intra prediction mode can be 19, when the size of block is 64 × 64, intra prediction mould
The quantity of formula can be 7.
The quantity of intra prediction mode can be fixed as N, without plumber block size how.For example, intra prediction mode
It is at least one of 35 or 67 that quantity, which can be fixed, without plumber block size how.
The quantity of intra prediction mode can according to color component type and change.For example, the quantity of prediction mode can be according to
It is luminance signal or carrier chrominance signal according to color component and changes.
Intraframe coding and/or solution can be executed by using sample value or coding parameter included in contiguous block is rebuild
Code.
In order to encode/decode according to intra prediction to current block, it can recognize to include the sampling point in reconstruction contiguous block
Whether the reference sampling point of coding/decoding object block can be used as.When in the presence of the reference sampling point that cannot act as coding/decoding object block
When sampling point, by using including at least one sampling point in the sampling point rebuild in contiguous block, sample value is replicated and/or interpolation
To the sampling point that cannot act as with reference to sampling point, it is thus impossible to be used as coding/decoding object block as the sampling point with reference to sampling point
With reference to sampling point.
In intra prediction, at least one of the size based on intra prediction mode and coding/decoding object block, filter
Wave device can be applied to reference at least one of sampling point or prediction sampling point.Here, coding/decoding object block can be meant currently
At least one of block, and can mean encoding block, prediction block and transform block.It is applied to reference to sampling point or predicts the filter of sampling point
The type of wave device can at least one of size/shape according to intra prediction mode or current block and change.The class of filter
Type can at least one of quantity, filter coefficient value or filter strength according to filter tap and change.
In the non-direction plane mode among intra prediction mode, when generating the prediction block of coding/decoding object block,
Can according to sampling point position by using on current sample with reference to the left upper right with reference to sampling point, current block of sampling point, current sample
The sample value in prediction block is generated with reference to the weighted sum of sampling point with reference to the lower-left of sampling point and current block.
It, can when generating the prediction block of coding/decoding object block in the non-direction DC mode among intra prediction mode
By generating prediction block with reference to the left mean value with reference to sampling point of sampling point and current block on current block.In addition, can be by using
With reference to sample value to one or more top rows adjacent with reference sampling point and an one or more left sides in coding/decoding block
Side column execute filtering.
In the case where multiple directions mode (angle mode) among intra prediction mode, it can be referred to by using upper right
Sampling point and/or lower-left generate prediction block with reference to sampling point, and the multiple direction mode can have different directions.In order to produce
Raw prediction sample value, can be performed the interpolation of real number unit.
It, can be pre- from the intra prediction mode of the neighbouring prediction block adjacent with current prediction block in order to execute intra-frame prediction method
Survey the intra prediction mode of current prediction block.It is predicted by using the pattern information predicted from neighbouring intra prediction mode
In the case where the intra prediction mode of current prediction block, in current prediction block and neighbouring prediction block intra prediction mould having the same
When formula, the information of current prediction block and neighbouring prediction block intra prediction mode having the same can be by using predetermined flag information
It is sent.When the intra prediction mode of current prediction block is different from the intra prediction mode of neighbouring prediction block, execution can be passed through
Entropy coding encodes the intraprediction mode information of coding/decoding object block.
Fig. 7 is the diagram for the embodiment for explaining the processing of inter-prediction.
Quadrangle shown in Fig. 7 can refer to diagram picture (or picture).In addition, the arrow of Fig. 7 can indicate prediction direction.Also
It is to say, image can be encoded or decode or be encoded according to prediction direction and decoding.According to type of coding, each image can
It is classified as I picture (picture in frame), P picture (single directional prediction picture), B picture (bi-directional predicted picture) etc..Each picture can
According to each picture type of coding and be encoded and decoding.
When the image as encoding target is I picture, image itself can be in the case where being not necessarily to inter-prediction by frame
Coding.When the image as encoding target is P picture, inter-prediction can be carried out by using the reference picture only on forward direction
Or motion compensation encodes image.It, can be by using in forward direction and inverse when the image as encoding target is B picture
Reference picture in the two carries out inter-prediction or motion compensation to encode to image.It selectively, can be by using
Forward direction and it is one of reverse on reference picture carry out inter-prediction or motion compensation image encoded.Here, work as frame
Between prediction mode by use, inter-prediction or motion compensation can be performed in encoder, and decoder may be in response to encoder and hold
Row motion compensation.Being encoded or decoded by using reference picture or is encoded can with the image of decoded P picture and B picture
It is considered as the image for inter-prediction.
Hereinafter, it will be described in inter-prediction according to the embodiment.
Inter-prediction or motion compensation can be executed by using both reference picture and motion information.In addition, interframe is pre-
Above-mentioned skip mode can be used in survey.
Reference picture can be at least one of previous picture and subsequent pic of current picture.Here, inter-prediction
The block of current picture can be predicted according to reference picture.Here, reference picture can mean the figure used when predicting block
Picture.Here, the region in reference picture can be by using reference picture indices (refIdx), the motion vector for indicating reference picture
Etc. indicating.
Reference block relevant to current block in inter-prediction selecting reference picture and reference picture.It can be by using selection
Reference block generate the prediction block of current block.Current block can be conduct present encoding target among the block of current picture or
The block of current decoding target.
Motion information can be derived from the processing of inter-prediction by encoding device 100 and decoding device 200.In addition, deriving
Motion information can be used when executing inter-prediction.Here, encoding device 100 and decoding device 200 can be by using weights
It builds the motion information of contiguous block or improves code efficiency or decoding efficiency with the motion information of position block (col block) or both of the above
Or both.Col block can be the space bit with coding/decoding object block in previously reconstructed same position picture (col picture)
Set relevant piece.Rebuilding contiguous block can be the block in current picture and passes through coding or decoding or coding and decoding two
The block that person had previously been reconstructed.In addition, reconstructed block can be the block adjacent with coding/decoding object block, or it is located at coding/solution
Block or both of the above at the outer corner of code object block.Here, at the outer corner of coding/decoding object block
Block can be with the horizontally adjacent contiguous block in coding/decoding object block vertically adjacent to block.Selectively, be located at coding/
Decode object block outer corner at block can be with vertically adjacent to the contiguous block in coding/decoding object block it is horizontally adjacent
Block.
Encoding device 100 and decoding device 200 can determine respectively be present in col picture with coding/decoding object block
Block at the position of space correlation, and predefined relative position can be determined based on determining block.It is described predefined opposite
Position can be present in the interior location of the block at the position of coding/decoding object block space correlation or external position or
Both interior location and external position.In addition, encoding device 100 and decoding device 200 can be based on identified described predefined
Relative position derive col block respectively.Here, col picture can be at least one for including in reference picture list
A picture in reference picture.
The method for deriving motion information can change according to the prediction mode of coding/decoding object block.For example, being applied to frame
Between the prediction mode predicted may include advanced motion vector prediction (AMVP), merging patterns etc..Here, merging patterns are referred to alternatively as
Move merging patterns.
For example, working as AMVP as prediction mode by application, encoding device 100 and decoding device 200 can be by using weights
The motion vector of contiguous block or the motion vector of col block or both of the above are built to generate motion vector candidates list respectively.Weight
The motion vector or both of the above of the motion vector or col block of building contiguous block are used as motion vector candidates.Here, col
The motion vector of block is referred to alternatively as temporal motion vector candidate, and the motion vector for rebuilding contiguous block is referred to alternatively as spatial motion vectors
It is candidate.
Encoding device 100 can produce bit stream, and bit stream may include motion vector candidates index.That is, coding is set
Standby 100 can generate bit stream by indexing progress entropy coding to motion vector candidates.Motion vector candidates index can indicate from
Including the optimal motion vectors candidates selected in the motion vector candidates in motion vector candidates list.Motion vector candidates
Index can be sent to decoding device 200 from encoding device 100 by bit stream.
Decoding device 200 can index motion vector candidates from bit stream and carry out entropy decoding, and can be by using by entropy
Selection decoding mesh among decoded motion vector candidates index motion vector candidates included in motion vector candidates list
Mark the motion vector candidates of block.
Encoding device 100 can calculate the motion vector difference between the motion vector and motion vector candidates of decoding object block
(MVD), and entropy coding can be carried out to MVD.Bit stream may include the MVD by entropy coding.MVD can be set by bit stream from coding
Standby 100 are sent to decoding device 200.Here, decoding device 200 can carry out entropy decoding to the MVD received from bit stream.Solution
Decoding apparatus 200 can derive the motion vector of decoding object block by the sum of decoded MVD and motion vector candidates.
Bit stream may include the reference picture indices etc. for indicating reference picture, and reference picture indices can be coded by entropy simultaneously
Decoding device 200 is sent to from encoding device 100 by bit stream.Decoding device 200 can be by using the movement of contiguous block
Information carrys out the motion vector of prediction decoding object block, and can be pushed away by using the motion vector and motion vector difference that predict
Lead the motion vector of decoding object block.Decoding device 200 can based on the motion vector and reference picture indices information derived come
Generate the prediction block of decoding object block.
As the another method for deriving motion information, merging patterns are used.Merging patterns can mean multiple pieces of movement
Merging.Merging patterns can mean that the motion information of a block is applied to another block.When merging patterns are by application, coding
Equipment 100 and decoding device 200 can by using the motion information or col block for rebuilding contiguous block motion information or more than two
Person generates merging candidate list respectively.Motion information may include at least one of following item: 1) motion vector, 2) refer to picture
Face index and 3) inter-prediction indicator.Prediction indicator can indicate unidirectional (L0 prediction, L1 prediction) or two-way.
Here, merging patterns can be applied to each CU or each PU.When merging patterns are held in each CU or each PU
When row, encoding device 100 can generate bit stream by carrying out entropy decoding to predefined information, and can send bit stream to
Decoding device 200.Bit stream may include the predefined information.The predefined information can include: 1) closed as instruction
And whether mode is directed to the merging mark for the information that each piece of subregion is performed, 2) as indicating the neighbour adjacent with encoding target block
Nearly block which of the merging index of information that is merged of block.For example, the contiguous block adjacent with encoding target block may include compiling
The code left contiguous block of object block, the upper contiguous block of encoding target block, time contiguous block of encoding target block etc..
Merging candidate list can indicate to store the list of motion information.In addition, mould can be merged executing by merging candidate list
It is generated before formula.Being stored in the motion information merged in candidate list can be at least one of following motion information: with
Same position block relevant to coding/decoding object block in the motion information of the adjacent contiguous block of coding/decoding object block, reference picture
Motion information, pass through to the preparatory combination for merging motion information present in motion candidates list and newly generated movement letter
Breath and zero merging are candidate.Here, the motion information of the contiguous block adjacent with coding/decoding object block is referred to alternatively as space conjunction
And it is candidate.It is candidate to be referred to alternatively as time merging for the motion information of same position block relevant to coding/decoding object block in reference picture.
Skip mode can be the mode that the pattern information of contiguous block itself is applied to coding/decoding object block.It skips
Mode can be one of the mode for inter-prediction.When skip mode is by use, encoding device 100 can be to about which block
The information of the motion information motion information that is used as encoding target block carry out entropy coding, and the information can be sent out by bit stream
It is sent to decoding device 200.Encoding device 100 can not send decoding device for other information (for example, syntax element information)
200.Syntax element information may include at least one of motion vector difference information, coded block flag and transform coefficient levels.
The residual signals generated after intra prediction or inter-prediction can pass through the change of a part as quantification treatment
It changes processing and is converted to frequency domain.Here, converting for the first time can be used DCT type 2 (DCT-II) and various DCT, DST cores.These
Transformation kernel can execute the separable transformation that 1D transformation is executed along horizontal and/or vertical to residual signals, or can be to residual error
Signal executes the inseparable transformation of 2D.
For example, DCT the and DST type used in transformation can be used as shown in following table in the case where 1D transformation
DCT-II, DCT-V, DCT-VIII, DST-I and DST-VII.For example, as Table 1 and Table 2 below, can derive and pass through composition
Mapping ensemble and DCT the or DST type used in transformation.
[table 1]
[table 2]
| Mapping ensemble | Transformation |
| 0 | DST_VII、DCT-VIII、DST-I |
| 1 | DST-VII、DST-I、DCT-VIII |
| 2 | DST-VII、DCT-V、DST-I |
For example, as shown in figure 8, according to intra prediction mode, different mapping ensembles for horizontally and vertically by
Definition.Next, encoder/decoder can be by using present encoding/decoding object block intra prediction mode and related change
The transformation of collection is changed to execute transformation and/or inverse transformation.In this case, entropy coding/decoding is not executed to mapping ensemble, and compiled
Code device/decoder can define mapping ensemble according to identical rule.In this case, which among the transformation of mapping ensemble is indicated
The entropy coding/decoding that kind transformation is used can be performed.For example, when the size of block is equal to or less than 64 × 64, according in frame
Prediction mode, three mapping ensembles as shown in table 2 are combined into, and three transformation are used for each horizontal direction transformation and vertical
Direction transformation merges with group executes a total of nine different transform method.Next, to believe residual error by using optimal transformation method
It number encodes/decodes, code efficiency can be enhanced as a result,.Here, among in order to be converted to three about a mapping ensemble
The information which transform method is used executes entropy coding/decoding, and truncation unitary binaryzation can be used.Here, in order to hang down
At least one of straight transformation and horizontal transformation, the information that can be used which of transformation for indicating mapping ensemble transformation execute
Entropy coding/decoding.
After completing above-mentioned transformation for the first time, as shown in Figure 9, encoder can execute secondary for the coefficient by transformation
Transformation is to improve encircled energy.The executable separable transformation that 1D transformation is executed along horizontal and/or vertical of quadratic transformation,
Or the executable inseparable transformation of 2D.Used information converting can be sent, or can be by encoder/decoder according to working as
Preceding encoded information and neighbouring encoded information are derived.For example, the mapping ensemble for quadratic transformation can be defined as 1D is converted.No
Entropy coding/decoding is executed to the mapping ensemble, and encoder/decoder can define mapping ensemble according to identical rule.This
In the case of, indicate the transformation of mapping ensemble which of the information that is used of transformation can be sent, and the information can pass through frame
Interior prediction or inter-prediction are applied to an at least residual signals.
It is different for each mapping ensemble to convert at least one of candidate quantity or type.Convert candidate quantity or
At least one of type can be determined differently based at least one of following item: the position of block (CU, PU, TU etc.),
Direction/non-direction of size, zoned format and prediction mode (mode within the frame/frames) or intra prediction mode.
Decoder can change whether be executed to execute the knot that secondary inverting is changed, and can be changed from secondary inverting according to secondary inverting
Whether fruit is executed to execute inverse transformation for the first time according to inverse transformation for the first time.
Above-mentioned transformation and quadratic transformation for the first time can be applied at least one signal component in brightness/chroma component, or
Person can be applied according to the size/shape of any encoding block.Entropy coding/decoding can be executed to the index of the following the two of instruction: first
Whether secondary transformation/quadratic transformation is used to and used transformation/quadratic transformation for the first time arbitrarily in encoding block.It may be selected
Ground, the index can be defaulted and be derived according at least one current/neighbouring encoded information by encoder/decoder.
The residual signals generated after intra prediction or inter-prediction are after transformation for the first time and/or quadratic transformation
By quantification treatment, and the transformation coefficient by quantization is handled by entropy coding.Here, as shown in Figure 10, by quantization
Transformation coefficient can at least one of the size/shape based on intra prediction mode or smallest blocks, according to diagonal direction, vertical
Direction and horizontal direction are scanned.
In addition, perform entropy decoding by quantization transformation coefficient can and being scanned in reverse according to block form by cloth
It sets, and at least one of inverse quantization or inverse transformation can be executed to related blocks.Here, as the method for inverse scan, diagonal side
It can be performed at least one of scanning, horizontal direction scanning and vertical scan direction.
For example, present encoding block size be 8 × 8 when, can to for 8 × 8 block residual signals execution become for the first time
Change, quadratic transformation and quantization, next, can at least one of three kinds of scanning sequency methods according to shown in Figure 10 be directed to
Each of four 4 × 4 sub-block executes scanning and entropy coding to the transformation coefficient by quantization.In addition, execution can be passed through
Entropy decoding to execute inverse scan to the transformation coefficient by quantization.The transformation coefficient by quantization for performing inverse scan is passing through
Become transformation coefficient after inverse quantization, and secondary inverting changes or at least one of inverse transformation is performed for the first time, rebuilds as a result,
Residual signals can be generated.
In Video coding processing, a block can be partitioned as shown in figure 11, and indicator corresponding with partition information
It can be sent with signal.Here, partition information can be at least one of following item: subregion sign (split_flag), four
Fork/binary tree mark (QB_flag), quaternary tree subregion sign (quadtree_flag), binary tree subregion sign
(binarytree_flag) and binary tree divisional type mark (Btype_flag).Here, split_flag is that instruction block is
The no mark being partitioned, QB_flag are that instruction block is the mark being partitioned according to quaternary tree form also according to binary tree form,
Quadtree_flag is the mark for indicating block and whether being partitioned according to quaternary tree form, and binarytree_flag is that instruction block is
The no mark being partitioned according to binary tree form, Btype_flag be indicate in the case where the subregion of binary tree form block be by
Vertical partitioning is still by the mark of horizontal partitioning.
When subregion sign is 1, it may indicate that subregion is performed, when subregion sign is 0, it may indicate that subregion is not held
Row.In the case where four forks/binary tree mark, 0 can indicate quaternary tree subregion, and 1 can indicate binary tree subregion.It may be selected
Ground, 0 can indicate binary tree subregion, and 1 can indicate quaternary tree subregion.In the case where binary tree divisional type mark, 0 can indicate water
Square vertical direction subregion can be indicated to subregion, 1.Selectively, 0 vertical direction subregion can be indicated, 1 can indicate horizontal direction point
Area.
For example, can be by will be in quadtree_flag, binarytree_flag and Btype_flag as shown in table 3
At least one the partition information for deriving Figure 11 is sent with signal.
[table 3]
For example, can be by will at least one of split_flag, QB_flag and Btype_flag as shown in table 4
The partition information for deriving Figure 11 is sent with signal.
[table 4]
Partition method only can be executed according to binary tree form according to quaternary tree form or only according to the size/shape of block.
In this case, split_flag can mean that instruction is to execute subregion also according to binary tree form according to quaternary tree form
Mark.The size/shape of block can be derived according to the depth information of block, and depth information can be sent with signal.
When the size of block is in preset range, subregion only can be executed according to quaternary tree form.Here, the preset range
Can be defined as only can be according at least one of the size for the largest block that quaternary tree form is partitioned or size of smallest blocks.
The information of largest block/smallest blocks size of the subregion of instruction permission quaternary tree form can be sent by bit stream signal, and
It and can be that unit sends the information with signal at least one of sequence, frame parameter or band (segmentation).Selectively,
Largest block/smallest blocks size can be the preset fixed dimension in encoder/decoder.For example, the model of the size when block
It encloses when being 256 × 256 to 64 × 64, only can execute subregion according to quaternary tree form.In this case, split_flag can be with
It is the mark for indicating whether to execute subregion according to quaternary tree form.
When the size of block is in preset range, subregion only can be executed according to binary tree form.Here, the preset range
Can be defined as only can be according at least one of the size for the largest block that binary tree form is partitioned or size of smallest blocks.
The information of largest block/smallest blocks size of the subregion of instruction permission binary tree form can be sent by bit stream signal, and
It and can be that unit sends the information with signal at least one of sequence, frame parameter or band (section).Selectively, most
Bulk/smallest blocks size can be the preset fixed dimension in encoder/decoder.For example, the range of the size when block
When being 16 × 16 to 8 × 8, subregion only can be executed according to binary tree form.In this case, split_flag can be instruction
Whether the mark of subregion is executed according to binary tree form.
It, can be only when the block that subregion goes out is by further subregion after carrying out subregion to a block according to binary tree form
Subregion is executed according to binary tree form.
When the width dimensions for the block that subregion goes out or length dimension cannot be by further subregions, at least one indicator can not
It is sent with signal.
Other than the binary tree subregion based on quaternary tree, the subregion based on quaternary tree can be held after binary tree subregion
Row.
When block is based on quaternary tree form or binary tree form or both and is partitioned, can will divide with according to the final of block
The corresponding block of the leaf node in area is set as single coding/decoding unit.In other words, when with arbitrary dimension or any form of piece
When being no longer partitioned, coding/decoding can be executed to corresponding block.In one embodiment, for arbitrary dimension or arbitrarily
Form and with pass through quaternary tree form subregion or binary tree form subregion or quaternary tree form subregion and binary tree form subregion
Prediction (for example, inter prediction or infra-frame prediction), transformation etc. can be performed in the corresponding block of y-bend leaf node that the two generates
Coding/decoding processing.
Figure 12 is the exemplary diagram for showing the coding/decoding unit according to the zoned format of block.It is shown in FIG. 12
In example, solid line is for distinguishing the block generated by quaternary tree subregion, and dotted line is for distinguishing the block generated by binary tree subregion.
It, can quilt by the node of solid line and the final subregion of dotted line when assuming that the structure of encoding block is confirmed as example shown in Figure 12
It is defined as y-bend leaf node.Can not according to prediction sub-block or transformation sub-block carry out additional partitions in the case where according to respective leaves
The corresponding block size of node or block form pair block corresponding with y-bend leaf node execute coding/decoding (for example, intra prediction or frame
Between prediction, for the first time transformation, quadratic transformation, quantization, entropy coding/decoding etc.).
For convenience of explanation, in the embodiment that will be described later, based on quaternary tree form or binary tree form or
The zoned format of the block of the two is defined as block structure.
When coding/decoding, the block structure of each color component can identical or each color component block structure can not
Together.In one embodiment, according to any coding parameter condition, block structure can be identical with chromatic component for luminance component, or
It can be for luminance component and chromatic component difference.Here, block structure is identical with chromatic component for luminance component indicates to be directed to
Luminance component determine block structure information by inherit arrive chromatic component, or for chromatic component determine block structure information by after
Hold luminance component.For example, luminance signal and carrier chrominance signal are in frame according to present encoding/decoded picture or the type of band
It can block structure having the same or different block structures in band in picture or in frame.It here, can be by according to each block structure
It derives rate distortion costs function and cost function is selected to become the cataloged procedure of the smallest block structure to determine configuration frame inside-paint
The luminance component of band and the block structure of chromatic component are set as identical or different in face or frame.
The information whether encoding device can be used for each color component to the identical block structure of expression carries out entropy coding, and
Transmit this information to decoding device.Here, which can be according to sequence grade (for example, sequence parameter set (SPS)), picture etc.
In grade (for example, parameter sets (PPS)), slice header, maximum coding unit (LCT or CTU) and coding unit (or encoding block)
At least one and be encoded.
For example, indicating that the block structure of each color component is directed to picture in frame, band, inter picture or interframe band in frame
Whether identical coding parameter information can be sent by SPS or PPS.
In addition, indicate the block structure of each color component in frame in band or interframe band whether identical coding parameter
Information can be sent by slice header.
In addition, indicating the block structure of each color component for maximum coding unit or the whether identical coding of coding unit
Parameter information can be sent as unit of maximum coding unit or coding unit.
When coding/decoding object block meets predetermined condition, can not allow to carry out block subregion to coding/decoding object block.
Therefore, the coding/decoding for meeting the block partition information of the block of predetermined condition can be omitted.Here, predetermined condition can with block size,
At least one of block form and block subregion depth are related, and can indicate to allow or do not allow quaternary tree form or binary tree form
Or size, form or the depth of the block of the subregion of two kinds of forms.Block size or form, which can be expression, to be allowed or does not allow four
Pitch the basic value of the size of the block of the subregion of tree-like formula or binary tree form or two kinds of forms, form or depth.Block depth can
Indicate the threshold value of the block depth for the subregion for allowing or not allowing quaternary tree form or binary tree form or two kinds of forms.Block depth
It can be according to the subregion for executing quaternary tree form or binary tree form or two kinds of forms with the 1 incremental factor.
Block partition information may include the information (for example, split_flag) for indicating whether perfoming block subregion, indicate whether to hold
The information (for example, Quadtree_flag or QB_flag) of row quaternary tree subregion indicates whether to execute the information of binary tree subregion
In the information (for example, Btype_flag) of (for example, Binarytree_flag or QB_flag) and expression binary tree divisional type
At least one.
For example, when assuming that predetermined condition indicates that block size is equal to or less than basic value and does not allow to meeting predetermined condition
Block carry out binary tree subregion when, for have equal to or less than basic value block size block, can omit and binary tree subregion
Related information is (for example, four forks/binary tree formal denotation (QB_flag), binary tree subregion sign (binaraytree_flag)
With binary tree divisional type mark (Btype_flag)) at least one of coding/decoding.When four forks of omission/binary tree
When the coding of formula mark (QB_flag), subregion sign (split_flag) can be used for indicating whether executing quaternary tree subregion to block.
It is not limited to above-mentioned example, it is settable to allow to carry out quaternary tree subregion to the block for meeting predetermined condition.Here, right
In the block for meeting predetermined condition, information related with quaternary tree subregion can be omitted (for example, four forks/binary tree formal denotation (QB_
Flag) or quaternary tree subregion sign (quadtree_flag)) at least one of coding/decoding.When four fork/bis- can be omitted
When the coding/decoding of fork tree formal denotation (QB_flag), subregion sign (split_flag) can be used for indicating whether executing block
Binary tree subregion.
In another embodiment, settable to allow to carry out any form of subregion to the block for meeting predetermined condition.Here,
For meeting the block of predetermined condition, any partition information can not be encoded/decoded.
Referring to attached drawing, will be described in determining whether to omit the processing to the coding/decoding of partition information.
Figure 13 is the flow chart for depicting a determination whether the processing being decoded to information related with binary tree subregion.In order to
It is convenient to explain, in this example, it is assumed that not allowing to carry out the subregion based on binary tree form to the block for meeting predetermined condition.
Firstly, can get information related with predetermined condition in step S1301.Here, information related with predetermined condition
It may include at least one of block size, block form and subregion depth.Predetermined condition can be set based on the information of predetermined condition
Whether it is equal to or less than whether threshold value, block form are presets, block depth for whether block size is equal to or more than threshold value, block size
Degree whether is equal to or more than threshold value or whether block depth is equal to or less than threshold value.
Information related with predetermined condition can be predefined in the encoder and the decoder.Here, have with predetermined condition
The information of pass can indicate to define at least one of block size, block form and the block depth of predetermined condition.In one embodiment,
Block size/the form or subregion depth that the coding/decoding of partition information is omitted can have in the encoder and the decoder in advance
The fixed value of definition.Selectively, information related with predetermined condition can pass through presentation code/decoding object block size/shape
The coding parameter of formula or the subregion depth of block is determined differently.
In another embodiment, information related with predetermined condition can be according to sequence grade, picture level, slice header or pre-
Yard territory element of delimiting the organizational structure is encoded/decoding.Here, predictive encoding region can have less than present encoding/decoded picture or band
Size/form, and may include maximum coding unit (LCU or CTU) or include in maximum coding unit have any ruler
Very little or any form of piece (for example, by executing the block that quaternary tree subregion generates to maximum coding unit).Have with predetermined condition
The information of pass can be expressed as the full-size of block or the minimum dimension of block or both, or can be expressed as the depth capacity or block of block
Minimum-depth or both.
Encoder can be lost by comparing the rate of result that encoded and be obtained based on quaternary tree form and binary tree form
The rate of result that is true and being encoded and obtained based on quaternary tree form is distorted to determine block structure.Encoder is contemplated that basis
Determining block structure no longer executes size, form or the depth of the block of binary tree subregion, to information related with predetermined condition into
Row coding.In addition, decoder can never allow the bit stream pair of binary tree subregion information related with predetermined condition to be decoded,
And determine whether current block meets predetermined condition based on decoded information.
In step S1302, decoder can determine whether current block meets predetermined condition.As a result, when current block meets
When predetermined condition, the decoding to information related with the binary tree subregion of current block can be omitted.
It selectively,, can be according to whether being executed to current block in step S1303 when current block is unsatisfactory for predetermined condition
Quaternary tree subregion is decoded information related with the binary tree subregion of current block.For example, four ought not be executed to current block
When fork tree subregion, information related with the binary tree subregion of current block can be decoded.
In other words, if to the block partition information of current block encode/decode can by comparing current block size,
Whether form or depth are corresponding to the size, form or depth according to the block of predetermined condition and are determined.
In another embodiment, according to an embodiment of the invention, indicating whether to allow to arbitrary dimension, arbitrary form
Or the information of the block progress block subregion of any depth can be encoded/decode.Here, the information for allowing to carry out block subregion is indicated whether
It may include indicating whether the information (for example, NoPresent_Quadtree_flag) of quaternary tree subregion or indicating whether to deposit
In the information (for example, NoPresent_Binarytree_flag) of binary tree subregion.
When instruction does not allow to carry out block subregion to the block with arbitrary dimension, arbitrary form or any depth, in addition to phase
It answers except block, can not allow to carry out block subregion to lower level block.Here, lower level block may include with the block ruler less than relevant block
Very little block, the block with identical with relevant block piece of form, the block with the subregion depth for being greater than relevant block and relevant block is more
At least one of low-level nodes block.
In one embodiment, when the letter for indicating whether to have to the binary tree subregion with arbitrary dimension/form block
Breath is transmitted with signal, and information instruction other than the block, is less than the block for having there is no when binary tree subregion
Size/form block, can omit to related with binary tree subregion information (for example, indicating whether to execute binary tree subregion
Information is (for example, four forks/binary tree mark (QB_flag), binary tree subregion sign (binaraytree_flag) and binary tree point
At least one of area's type code (Btype_flag)) coding/decoding.
Be not limited to above-mentioned example, indicate whether exist to arbitrary dimension/form block quaternary tree subregion and whether
There are the information of binary tree divisional type mark to be transmitted with signal.
Indicate whether that the information for allowing block subregion can be sent according to predictive encoding region.Here, predictive encoding region can
With being less than size/form of present encoding/decoded picture or band, and may include in maximum coding unit (LCU or CTU) or
Coding unit (for example, pass through to maximum coding unit execute quaternary tree subregion generate block) in include have arbitrary dimension or
Any form of piece.Encoder can be by comparing based on quaternary tree form and binary tree form to arbitrary dimension/form
Block encoded and the distortion of the rate of the result that obtains and encoded based on quaternary tree form and the rate distortion of result that obtains
To determine block structure, and determine whether that the information to expression permission binary tree subregion encodes according to determining block structure.
Indicate whether to allow the information of binary tree subregion can be encoded/decoding according to layer.In one embodiment, when with letter
When the information of number higher block sent indicates to allow block subregion, indicate whether to allow to being generated by subregion higher block
The information that lower level block carries out block subregion can be encoded/decode.
In another embodiment, the size, form of block or the information of depth (indicate whether perfoming block subregion in the information
Information transmitted with signal) can according to greater degree be encoded/decoding.In one embodiment, the size, form or depth of block
The information of degree can be sent according at least one of sequence grade, picture level and slice header.Here, for by higher
Size, form or the corresponding block of depth for the block that grade is transmitted with signal, or for the higher block in the greater degree,
Indicate whether that the information for allowing block subregion can be transmitted with signal.
Figure 14 is the flow chart for depicting a determination whether the processing being decoded to information related with binary tree subregion.In order to
It is convenient to explain, in this example, it is assumed that indicating whether binary tree subregion is only allowed for the information of current block to be passed with signal
It send.
Firstly, can be decoded to the information for executing binary tree subregion is indicated whether in step S1401.
The binary tree point to current block can be omitted when information expression does not allow to carry out binary tree subregion in step S1402
The decoding of area's information.In addition, can not be to the binary tree point for the lower level block by being generated by the current block of quaternary tree subregion
Area's information is decoded.
It,, can be according to being in step S1403 when information expression allows to carry out binary tree subregion meanwhile in step S1402
It is no that quaternary tree subregion is executed to current block to be decoded to information related with binary tree subregion.For example, ought not be to current block
When executing quaternary tree subregion, information related with the binary tree subregion of current block is directed to can be decoded.In addition, for passing through
The lower level block for executing quaternary tree form or binary tree subregion to current block and generating, can be according to whether execute four to lower level block
Fork sets subregion to be decoded to information related with binary tree subregion.
Figure 15 to Figure 17 is shown no longer for the feelings with predetermined size or smaller size of piece of execution binary tree subregion
The exemplary diagram of condition.
As example shown in Figure 15, it is assumed that size/form of maximum coding unit is 128 × 128, does not execute y-bend
Subregion is set, only quaternary tree subregion is present in maximum coding unit by the rate-distortion optimization executed by encoding device.
Shown in example as shown in Figure 16, when the information that expression does not execute binary tree subregion to predetermined size block is not compiled
When code/decoding, indicate whether that the information that binary tree subregion is executed to the block for no longer executing quaternary tree subregion can be encoded/decode.
However, as shown in Figure 17 shown in example, when expression is held to 128 × 128 sizes or smaller size of piece
When the information of row binary tree subregion is encoded/decodes, indicate whether to 128 × 128 sizes or smaller size of piece of execution
The information of binary tree subregion can not be encoded/decode.Therefore, information content encoded is reduced, so that coding/decoding efficiency mentions
It is high.
Above with reference to described in Figure 13, encoder can size to the block for not allowing to carry out binary tree subregion (for example, indicating
128 × 128 information), the information of form or depth encoded, and sends decoding device for the information of coding.Decoding is set
It is standby to be decoded from information of the bit stream to the block size for not executing binary tree subregion, and no longer to being equal to or less than with having
The related information of binary tree subregion of the block of the block size of the size indicated by decoded information is decoded.
In another embodiment, above with reference to described in Figure 14, encoding device can not allow to arbitrary dimension expression
Block (block does not execute binary tree subregion) carry out the information of binary tree subregion and encoded, and send solution for the information of coding
Decoding apparatus.Here, information can be 1 bit flag (for example, NoPresent_BinaryTree_flag), but not limited to this.?
In example shown in Figure 17, for example, the NoPresent_BinaryTree_flag of the block with 128 × 128 sizes is by with letter
Number send.
In Figure 16 and Figure 17, for example, value of statistical indicant is arranged to 1 when quaternary tree form or binary tree subregion are performed,
Otherwise, value of statistical indicant is arranged to 0.However, opposite setting is also possible.
Embodiment related with not allowing to carry out block subregion can be applied to luminance component and chromatic component.Here, it indicates
Do not allow to carry out block subregion information (for example, indicate the information of the size of block for not allowing to carry out block subregion, form or depth,
Perhaps indicate whether to allow to carry out the information of block subregion) it can be universally applied to luminance component and chromatic component or can needle
Luminance component and chromatic component are independently transmitted with signal.When information is coded by entropy/decodes, this binaryzation side of Lay is truncated
Method, K rank Exp-Golomb binarization method, limitation K rank Exp-Golomb binarization method, fixed length binarization method, unitary two
Any one in value method and truncation unitary binarization method is used as entropy coding method.In addition, being carried out to information
After binaryzation, finally information is encoded/decoded by using CABAC (ae (v)).
Next, by the transformation and scanning of the residual signals for describing current block.
When the residual signals to current block encode/decode, the coding/decoding block adjacent with current block can be passed through
The encoded information of residual signals implicitly derives at least one coding letter of the residual signals of current block in encoder/decoder
Breath.Here, the encoded information of residual signals may include with the conversion scheme of residual signals (for example, for for the first time transformation and it is secondary
The conversion scheme of transformation) related information and the transformation coefficient for scanning quantization information.Here, the transformation coefficient of quantization can
It indicates to perform the residual signals generated after intra prediction transformation (for example, transformation and quadratic transformation for the first time) and quantization.
In detail, when current block is encoded by intra prediction, can the intra prediction mode based on current block from work as
Preceding piece of adjacent contiguous block derives the encoded information of current block.It selectively, can when current block is encoded by inter-prediction
The encoded information of current block is derived from the contiguous block adjacent with current block based on the motion information of current block.Hereinafter, referring to Fig.1 8
And Figure 19, it will be described in being encoded when current block by intra prediction and current block pass through inter-prediction Shi Conglin encoded
Nearly block derives the processing of the encoded information of the residual signals of current block.
Figure 18 is to show to determine whether to derive the residual of current block from contiguous block when current block is encoded by intra prediction
The flow chart of the processing of the encoded information of difference signal.
Firstly, in step S1801, it may be determined whether exist according to pre- in frame identical with the intra prediction mode of current block
Survey mode contiguous block encoded.Here, (in other words, the contiguous block of current block can be included in identical picture with current block
Current picture) in, and indicate to be encoded/decoded piece before current block.In one embodiment, contiguous block may include working as
Be encoded before preceding piece/decoded piece in the block adjacent with current block.Here, the block adjacent with current block may include with currently
The adjacent block in the boundary (for example, left border or boundary) of block and with the turning of current block (for example, upper left corner, upper right
Turning or lower-left corner) adjacent at least one of block.
When existing according to intra prediction mode identical with the intra prediction mode of current block contiguous block encoded,
The encoded information for being used for the residual signals of corresponding contiguous block can be derived as the encoded information of current block by step S1802.In detail
Ground can derive the change for the first time of current block from the contiguous block with intra prediction mode identical with the intra prediction mode of current block
It changes, at least one of quadratic transformation and scanning information.
In one embodiment, when the intra prediction mode phase of the intra prediction mode of current block and the contiguous block of current block
When same and corresponding contiguous block has skipped transformation (transformation is skipped) for the first time, the residual signals of present encoding block also be can skip for the first time
Transformation.When the transformation for the first time of current block is skipped, the quadratic transformation of current block can also be skipped.
Selectively, when the intra prediction mode of current block is identical as the intra prediction mode of the contiguous block of current block,
Current block can be arranged to and be applied to the intra prediction for having with current block for transformation for the first time both horizontally and vertically
The transformation for the first time of the contiguous block of the identical intra prediction mode of mode is identical.Therefore, can omit to the residual signals of current block into
Row converts required encoded information (for example, the used change when be directed to transformation for the first time both horizontally and vertically for the first time
Change information (or manipulative indexing)) coding/decoding.
For example, the intra prediction mode when current block is confirmed as number 23 (mode 23), and adjacent with current block
When the intra prediction mode of at least one contiguous block is confirmed as number 23 (mode 23), applied to pre- in the frame with number 23
The transformation for the first time of the residual signals of the contiguous block of survey mode is used as the transformation for the first time of the residual signals for current block.Example
Such as, when for intra prediction mode identical with the intra prediction mode of current block contiguous block residual signals according to
The transformation for the first time of horizontal direction is performed by DCT-V, and logical for the transformation for the first time according to vertical direction of residual signals
It crosses DST-VII to be performed, the change for the first time according to horizontal direction of the residual signals for current block is executed by using DCT-V
It changes, and the transformation for the first time according to vertical direction is executed by using DST-VII.
In another embodiment, when the intra prediction mode phase of the intra prediction mode of current block and the contiguous block of current block
Meanwhile the quadratic transformation of current block can be arranged to and be applied to have in frame identical with the intra prediction mode of current block in advance
The quadratic transformation of the contiguous block of survey mode is identical.Therefore, it can omit and the residual signals of current block are executed needed for quadratic transformation
The coding/decoding of encoded information (for example, information converting (or manipulative indexing) of quadratic transformation).
For example, the intra prediction mode when current block is confirmed as number 35 (mode 35), and adjacent with current block
When the intra prediction mode of at least one contiguous block is also determined as number 35 (mode 35), applied in the frame with number 35
The quadratic transformation of the residual signals of the contiguous block of prediction mode is used as the quadratic transformation of the residual signals for current block.
In another embodiment, when the intra prediction mode phase of the intra prediction mode of current block and the contiguous block of current block
Meanwhile the scanning sequency of current block can be arranged to and have intra prediction mode identical with the intra prediction mode of current block
Contiguous block scanning sequency it is identical.Therefore, can omit to the transformation coefficient after the quantization for the residual signals of current block into
Encoded information needed for row scanning (for example, diagonal direction, horizontally and vertically in (indicate scanning sequency) at least one
A scanning index (scanning sequency index)) coding/decoding.
It is not limited to above-mentioned example, the contiguous block with intra prediction mode identical with the intra prediction mode of current block
Transformation for the first time, at least two in the quadratic transformation and scanning sequency encoded informations that can be derived as current block.
In one embodiment, there is the contiguous block of intra prediction mode identical with the intra prediction mode of current block
Two converted for the first time with scanning sequency or contiguous block that current block or contiguous block can be applied to quadratic transformation are converted for the first time
Secondary transformation and scanning sequency can be applied to current block.Selectively, there is frame identical with the intra prediction mode of current block
Current block can be all applied in transformation, quadratic transformation and the scanning sequency for the first time of the contiguous block of inner estimation mode.
When adjacent with current block with multiple contiguous blocks of current block intra prediction mode having the same, it can be based on neighbouring
Priority between block derives the encoded information of current block.In one embodiment, when the block adjacent with the left side of current block and
The block adjacent with the upside of current block is respectively provided with intra prediction mode identical with the intra prediction mode of current block, and with
The priority of the adjacent block in the left side of current block be higher than the block adjacent with the upside of current block priority when, can based on it is current
The encoded information of the adjacent block in the left side of block derives the encoded information of current block.
In another embodiment, as multiple neighbours with intra prediction mode identical with the intra prediction mode of current block
When nearly block is adjacent with current block, it can send for identification that (contiguous block is for deriving currently for contiguous block by bit stream signal
The encoded information of block) information.Here, the neighbour that can be indicated from the information (for example, contiguous block index) by contiguous block for identification
Nearly block derives the encoded information of the residual signals of current block.
When there is no the contiguous block with intra prediction mode identical with the intra prediction mode of current block, in step
S1803 can carry out entropy coding/decoding to the encoded information of the residual signals of current block.In one embodiment, when there is no tools
It, can be to the change of current block converted for the first time when having the contiguous block of intra prediction mode identical with the intra prediction mode of current block
Change the information (or scanning index) of information (or manipulative indexing), the information converting (or manipulative indexing) of quadratic transformation and scanning sequency
At least one of carry out entropy coding/decoding.
In the above-described embodiment, identical as the intra prediction mode of contiguous block simultaneously in the intra prediction mode of current block
And it is described in the case where the encoded information of the residual signals from contiguous block derivation current block.In another embodiment, may be used
Current block is derived from the contiguous block of the first encoded information with residual signals identical with the first encoded information of current block
Second encoded information of residual signals.Here, the first encoded information and the second encoded information may include the information converted for the first time, two
At least one of information and scanning sequency of secondary transformation.
In one embodiment, when exist using with for current block determine convert for the first time it is identical convert for the first time to
When a few contiguous block, the quadratic transformation of current block can be arranged to be applied to use head identical with the transformation for the first time of current block
The quadratic transformation of the contiguous block of secondary transformation.Here, volume needed for the residual signals execution quadratic transformation for current block can be omitted
The coding/decoding of code information.For example, it is assumed that being directed to being determined according to the transformation for the first time of horizontal direction for the residual signals of current block
For DCT-V, it is confirmed as DST-VII according to the transformation for the first time of vertical direction.When DCT-V is confirmed as at least one of current block
Contiguous block according to horizontal direction for the first time transformation and DST-VII be confirmed as the transformation for the first time according to vertical direction when, answer
The secondary of current block can be applied to the quadratic transformation that converts the identical contiguous block converted for the first time for the first time with current block
Transformation.
In addition, can be applied to using the scanning sequency for converting the identical contiguous block converted for the first time for the first time with current block
The scanning sequency of current block.In addition, using the quadratic transformation for converting the identical contiguous block converted for the first time for the first time with current block
The quadratic transformation and scanning sequency of current block can also be applied to scanning sequency.
In the above-described embodiment, it has been described that from use with current block convert for the first time it is identical convert for the first time it is neighbouring
Block derives at least one of quadratic transformation and scanning sequency of current block.However, can be from the quadratic transformation used with current block
The contiguous block of identical quadratic transformation derives at least one of the transformation and scanning sequency for the first time of current block, or can be from using
The contiguous block of scanning sequency identical with the scanning sequency of current block derives in the transformation for the first time and quadratic transformation of current block extremely
It is one few.
It can be from intra prediction mode identical with the intra prediction mode of current block and the first encoded information and first
The contiguous block of encoded information derives the second encoded information of current block.
In one embodiment, when exist using with for current block determine intra prediction mode and for current block it is true
When fixed at least one contiguous block for converting identical intra prediction mode for the first time and convert for the first time, the quadratic transformation of current block can
It is arranged to be applied to intra prediction mode identical with the intra prediction mode of current block and using the head with current block
The secondary quadratic transformation for converting the identical contiguous block converted for the first time.Here, the residual signals that can be omitted for current block execute two
The coding/decoding of encoded information needed for secondary transformation.
In addition, using the intra prediction mode with current block and converting identical intra prediction mode for the first time and converting for the first time
The scanning sequency of contiguous block can be applied to the scanning sequency of current block.Selectively, using the intra prediction with current block
Mode and the quadratic transformation of identical intra prediction mode and the contiguous block converted for the first time and scanning sequency are converted for the first time can be answered
For current block.
In the above-described embodiment, from have intra prediction mode identical with current block and using with current block for the first time
It converts the identical contiguous block converted for the first time and derives at least one of quadratic transformation and scanning sequency of current block.Except this it
It outside, can be from the quadratic transformation phase with the identical intra prediction mode with the intra prediction mode of current block and use and current block
With quadratic transformation contiguous block derive current block at least one of transformation and scanning sequency for the first time, or can from have with
The identical intra prediction mode of the intra prediction mode of current block simultaneously uses scanning sequency identical with the scanning sequency of current block
Contiguous block derive current block at least one of transformation and quadratic transformation for the first time.
Figure 19 is to show to determine whether to derive the residual of current block from contiguous block when current block is encoded by inter-prediction
The flow chart of the processing of the encoded information of difference signal.
Firstly, in step S1901, it may be determined that whether the inter-frame forecast mode of current block is merging patterns.When current block
When inter-frame forecast mode is merging patterns, in order to derive the motion information of current block, in step S1902, it may be determined that with current block
Combined contiguous block.In one embodiment, can by indicate to merge in candidate list by the contiguous block with current merged block
Merging index to determine the contiguous block with current merged block.Here, the contiguous block of current block may include and current block space phase
Adjacent contiguous block and the contiguous block temporally adjacent with current block.
It, can be by the contiguous block with current merged block in step S1903 when the contiguous block with current merged block is determined
The encoded information of residual signals is derived as the encoded information of the residual signals of current block.In one embodiment, the head of current block
At least one of secondary transformation, quadratic transformation and scanning sequency can be uniformly set as the head of the contiguous block with current merged block
At least one of secondary transformation, quadratic transformation and scanning sequency.
When the inter-frame forecast mode of current block is not merging patterns, in step S1904, it may be determined that in the neighbouring of current block
With the presence or absence of the contiguous block with motion information identical with the motion information of current block in block.Here, motion information may include
At least one of motion vector, reference picture indices and reference picture direction.
When presence has the contiguous block of motion information identical with the motion information of current block, in step S1905, can incite somebody to action
The encoded information of the residual signals of contiguous block with motion information identical with the motion information of current block is derived as current block
Residual signals encoded information.In one embodiment, in transformation, quadratic transformation and the scanning sequency for the first time of current block extremely
Few one can be uniformly set for in motion vector, reference picture indices and the reference picture direction with current block extremely
The change for the first time of the contiguous block at least one of a few identical motion vector, reference picture indices and reference picture direction
It changes, at least one of quadratic transformation and scanning sequency.
It, can in step S1906 when there is no the contiguous block with motion information identical with the motion information of current block
Entropy coding/decoding is carried out to the encoded information of the residual signals of current block.In one embodiment, when there is no have with it is current
When the contiguous block of the identical motion information of the motion information of block, can to current block convert for the first time information converting (or transformation rope
Draw), at least one of the information (or scanning index) of the information converting (or manipulative indexing) of quadratic transformation and scanning sequency into
Row entropy coding/decoding.
In the example being shown in FIG. 19, can according to the inter-frame forecast mode of current block whether be merging patterns adaptively
Determine the contiguous block for deriving the encoded information of the residual signals of current block.It is different from example shown in Figure 19, when current
When the inter-frame forecast mode of block is only merging patterns, the encoded information of the residual signals of current block can be derived from contiguous block.It is optional
Ground is selected, whether is that merging patterns are unrelated with the inter-frame forecast mode of current block, it can be identical as the motion information of current block from having
Motion information contiguous block derive current block encoded information.
In the above-described embodiment, identical as the motion vector of contiguous block in the motion vector of current block and from neighbouring
It is described in the case where the encoded information of the residual signals of block derivation current block.In another embodiment, can from have with
The contiguous block of identical first encoded information of the first encoded information or motion vector of the residual signals of current block or motion vector
Derive the second encoded information of the residual signals of current block.
Such as above-described embodiment derive current block motion information after, can the motion information based on current block whether with neighbour
The motion information of nearly block is identical to derive the encoded information of current block from contiguous block.In addition, can be based on the motion information of contiguous block
Derive motion information of the encoded information of current block without considering current block.
It can be based on indicating whether using predefined type (for example, predefined alternative types or predefined scan type) and being
No use is any one in the type of residual error (for example, real transform type or residual error scan type) other than predefined type
At least one a information encode/solve to above-mentioned encoded information (such as transformation, quadratic transformation and scanning sequency for the first time)
Code.
In one embodiment, when residual signals are generated by intra prediction or inter-prediction or both, table
Show whether predefined alternative types are applied to the information of residual signals and can be encoded.Here, predefined alternative types can
To be master's alternative types to be used (for example, DCT-II) when executing transformation for residual signals, but not limited to this.Information can
To be 1 bit flag (for example, transformation mark, TM mark).In one embodiment, it when TM mark is 0 (or 1), can indicate
Predefined alternative types are applied to residual signals.When TM mark is 1 (or 0), can indicate in addition to predefined transformation classes
Other alternative types except type are applied to residual signals.Furthermore, it is possible to be matched with having the mark of 2 bits or more
Confidence breath, the first bit can indicate whether predefined alternative types be used to convert for the first time, and the second bit can indicate predefined
Alternative types whether be used for quadratic transformation.
When information indicates that other alternative types other than predefined alternative types are applied to residual signals, use
The information of any one real transform type in specified real transform type can be encoded.Here, real transform type can
Indicate remaining alternative types that can be applied in the alternative types of residual signals other than predefined alternative types.Example
Such as, when predefined alternative types are DCT-II, real transform type may include DCT-V, DCT-VIII, DST-I and DST-
At least one of VII.Information can be the index information of any one real transform type in specified real transform type
(TM idx), and index information can be any positive integer.For example, TM idx 1 can indicate that DCT-V, TM idx 2 can be indicated
DCT-VIII, TM idx 3 can indicate that DST-I, TM idx 4 can indicate DST-VII.
Index information can indicate the alternative types combination horizontally and vertically for residual signals.In other words
It says, the 1D alternative types for horizontal can be determined by single index information.For example, when TM mark be 1 it is same
When TM idx when being 1, combining with the matched alternative types of TM idx 1 can be confirmed as the horizontal direction of current block and vertical
Histogram to alternative types.In one embodiment, when TM idx instruction for the DCT-V of horizontal direction and is indicated for vertical
When the DCT-VIII in direction, DCT-V and DCT-VIII can be determined respectively to be the horizontal direction alternative types of current block and vertical
Direction transformation type.
When determining the coding parameter of current block, any one specified in following item can be derived from the contiguous block of current block
At least one information: whether predefined type is used and type of residual error.For example, can be pushed away from the contiguous block of current block
Lead at least one information (TM idx) for specifying any one in following item: indicate predefined alternative types whether by
Information (TM mark) and real transform type applied to current block.
In one embodiment, at least one of the TM of current block mark and TM idx can be derived as and current block
The identical value of the value of contiguous block.
It, can be by implicitly assuming the TM of current block in addition, when at least one TM mark of the contiguous block of current block is 1
Mark is 1 to execute coding/decoding.Here, the TM idx of current block can implicitly be sent by bit stream, or can be from neighbour
Nearly block is implicitly derived.
It is described, can be derived from the contiguous block used when executing intra prediction or inter-prediction to current block with example
For specifying at least one information (TM idx) of any one in following item: indicating whether predefined alternative types are answered
Information (TM mark) and real transform type for current block.
In one embodiment, when the inter-frame forecast mode of current block is merging patterns, it is contemplated that TM mark and TM
It is candidate that at least one of idx newly to configure merging.The merging candidate list newly configured may include in TM mark and TM idx
At least one merging with different value is candidate.In one embodiment, merging candidate list can be configured to have first to close
And candidate and the second merging is candidate, wherein the first merging is candidate and the second merging is candidate with same motion information and with not
With TM mark or both TM idx or difference TM mark and TM idx.At least one of the TM mark of current block and TM idx
It can be confirmed as identical as at least one of the merging candidate by merging index (Merge_idx) instruction.It therefore, can be based on conjunction
And mode indicates the motion information (motion vector, reference picture indices, inter-prediction direction indicator) and TM of current block
Or both TM idx or TM mark and TM idx encode/decode.
Here, indicate that merging the information that candidate list is newly configured can explicitly be sent by bit stream.The letter of transmission
Breath can be 1 bit flag, but not limited to this.In addition, when the TM mark of at least one contiguous block of current block is 1, it can be hidden
Identify that merging candidate list is newly configured likes.Here, contiguous block can be first according to block scan sequence TM mark is predetermined proximity to
First become 1 block, or can be predefined block of locations.
In the above-described embodiment, it has been described that derive the information for determining alternative types from the contiguous block of current block
The method of (for example, TM mark or both TM idx or TM mark and TM idx).The embodiment can be applied to determination and work as
At least one of preceding piece of the alternative types converted for the first time and the alternative types for determining quadratic transformation determine.Implement at one
In example, in other words, the information converting converted for the first time can be derived from the contiguous block of current block (for example, TM mark (the first TM mark
Will) or TM idx (the first TM idx) or TM mark and both TM idx) and quadratic transformation information converting (for example, TM mark
At least one of will (the 2nd TM mark) or TM idx (the 2nd TM idx)).
In addition, can be produced based on the information converting of current block other than the merging candidate list generated based on motion information
Intercrescence and candidate list.In one embodiment, when the merging candidate list that the motion information based on contiguous block generates is defined
For " first merge candidate list ", and the merging candidate list that the information converting based on contiguous block generates is defined as " second
Merge candidate list " when, the merging candidate for indexing and specifying that merged by first merged in candidate list from first derives current block
Motion information.However, the merging candidate derivation specified by the second merging index that can merge in candidate list from second is current
The information converting of block.
In addition to this, the information for determining the scanning sequency of current block from the contiguous block of current block can be derived (for example, sweeping
Retouch both mark or scanning idx or scanning flag and scanning idx).Here, scanning flag can indicate the scanning sequency of current block
Whether identical as predefined scanning sequency, scanning idx can be any one residual error scanning in instruction residual error scanning sequency
The information of sequence.
According to another embodiment of the present invention, identical encoded information can be applied to all pieces be located in signaling block,
In, the signaling block is in present encoding/decoded picture or band.Here, signaling block can indicate have less than current picture or work as
The region of the size of the horizontal resolution or at least one of vertical resolution of preceding article band.In other words, signaling block can be determined
Justice is the presumptive area with the size for being less than current picture or current band.
The information of signaling block can be sent by least one of sequence units, picture unit and slice header.In a reality
Apply in example, can be sent by least one of sequence parameter set, parameter sets and slice header the size of signaling block, form or
At least one of position.It selectively, can be by the encoded information of current block or the contiguous block adjacent with current block implicitly
Derive the information of signaling block.Signaling block can have square or rectangular in form, but not limited to this.
The encoded information of signaling block can be applied to include all pieces in signaling block.In one embodiment, for the first time
At least one of transformation, quadratic transformation and scanning sequency can be uniformly set to including all pieces in signaling block.It can
By bit stream sending application in including all pieces of the encoded information in signaling block.Selectively, specific in signaling block
The encoded information of block of locations can be applied to include all pieces in signaling block.
In the above-described embodiment, it has been described that including all pieces of encoded informations having the same in signaling block.?
In another embodiment, the block that predetermined condition is met in the block that includes in signaling block can be arranged to coding letter having the same
Breath.Here, predetermined condition can be defined according at least one of size, form or depth of block.In one embodiment, for the first time
At least one of transformation, quadratic transformation and scanning sequency can be had by being uniformly set in all pieces for including in signaling block
There is predetermined size or smaller size of piece (for example, have 4 × 4 sizes or smaller size of piece).
The embodiment for obtaining the encoded information of current block can be applied to luminance component and chromatic component.In addition, logical
At least one of cross using described embodiment, indicate the residual signals for current block execute transformation for the first time, quadratic transformation and
The information of at least one of scanning can be encoded/decode.When above- mentioned information are coded by entropy/decode, this binaryzation of Lay is truncated
Method, K rank Exp-Golomb binarization method, limitation K rank Exp-Golomb binarization method, fixed length binarization method, unitary
At least one of binarization method and truncation unitary binarization method are used as entropy coding method.In addition, to above-mentioned letter
After breath carries out binaryzation, finally above- mentioned information can be encoded/decoded by using CABAC (ae (v)).Selectively,
The encoded information that determines current block can be implicitly derived by using at least one of the size of current block and form.
Next, will be described in encoding/decoding motion vector information.
When current block is encoded by inter-prediction, encoder can will indicate to be adjacent to movement encoded with current block
The motion vector difference (MVD) of difference between the motion vector of vector sum current block is sent to decoder.
Decoder can will be adjacent to the motion vector candidates that motion vector encoded is derived as current block with current block.In detail
Carefully, decoder can be from both the decoded temporal motion vector of current block and decoded spatial motion vectors or current block
At least one of decoded temporal motion vector or decoded spatial motion vectors derive motion vector candidates, and configure movement
Vectors candidates list (MVP list).
Encoder is transmittable to be indicated to include in the motion vector candidates in motion vector candidates list for deriving fortune
The information of the information (for example, MVP list index) of the motion vector predictor of dynamic vector difference.Then, decoding device can will be by MVP
The motion vector candidates of list index instruction are determined as motion vector predictor, and by using motion vector predictor and movement
The motion vector of phasor difference derivation current block.
Based on above-mentioned explanation, it will be described in the motion vector information according to the present invention to current block and encode/solve
The method of code.
Figure 20 is the flow chart for showing the decoding process of motion vector of current block.
Firstly, the spatial motion vectors that can derive current block are candidate in step S2001.It can be from being included in and include current
The spatial motion vectors that coding/decoding block in the identical picture of the picture of block derives current block are candidate.
Figure 21 is to show the exemplary diagram for deriving spatial motion vectors candidate.
Example as shown in Figure 21, can be from the block B1 adjacent with the upside of current block X, adjacent with the left side of current block
Block A1, the block B0 adjacent with the upper right corner of current block, block B2 and the block A0 adjacent with the lower left corner of current block positioned at the upper left corner
Derive the spatial motion vectors of current block.The spatial motion vectors derived from the contiguous block of current block can be confirmed as current block
Spatial motion vectors are candidate.
Here, spatial motion vectors candidate can be derived according to predetermined order.In one embodiment, spatial motion vectors are waited
Choosing can be determined according to the sequence of A0, A1, B0, B1 and B2 whether there is motion vector in each piece.When there are the fortune of contiguous block
When dynamic vector, the motion vector of corresponding contiguous block can be confirmed as spatial motion vectors candidate.
When the reference picture difference of the reference picture of contiguous block and current block, by using the ginseng referred to by current picture
Examine the distance between picture and contiguous block and by the distance between the reference picture of current picture reference and current block to contiguous block
Motion vector zoom in and out and the motion vector that obtains can be confirmed as the spatial motion vectors of current block.
Then, in step S2002, the temporal motion vector that can derive current block is candidate.It can be from the reconstruction in same position picture
The temporal motion vector of block derivation current block.
Figure 22 is to show the exemplary diagram for deriving temporal motion vector candidate.
Example as shown in Figure 22, can be from the block of the position H for the outside for being present in same position block C or from being present in same position block
The block of the position C3 of the inside of C, derives the temporal motion vector of current block, wherein the same position of current picture is corresponded to position block C
Position identical with the space current block X in picture.Time movement arrow can be sequentially derived from the block of the position H and the block of the position C3
Amount is candidate.In one embodiment, when motion vector can be derived from the block of the position H, time fortune can be derived from the block of the position H
Dynamic vector is candidate.Selectively, when motion vector can not be derived from the block of the position H, the time can be derived from the block of the position C3
Motion vector candidates.When the block of the position H or the position C3 is encoded by intra prediction, the time fortune of current block will not be derived
Dynamic vector is candidate.
Other than the example shown in Figure 22, the same position picture that can be indicated from the motion information of the current block by obtaining with
And by the same position block for including in the same position picture of motion information instruction or with the contiguous block of position block, derive at least one of current block
Temporal motion vector is candidate.Here, motion information may include the picture index for indicating same position picture and instruction in the picture of position
With at least one of the motion vector of position block.For specifying with position picture and current block quilt can be directed to the motion information of position block
In addition it is transmitted with signal.
The temporal motion vector that current block can be obtained according to the sub-block unit with the size less than current block is candidate.Example
It such as, can be according to the size (such as, 2 × 2,4 × 4,8 × 4,4 × 8 smaller than current block when the size of current block is 8 × 8
Deng) sub-block unit to obtain temporal motion vector candidate.Sub-block can have square or rectangular in form.In addition, the size of sub-block
Or form can be predetermined in encoder/decoder, or can be determined according to the size or form of current block.
Then, in step S2003, can produce including spatial motion vectors candidate and temporal motion vector candidate at least
The motion vector candidates list of one motion vector candidates.
Here, motion vector candidates list can be configured to include at least one temporal motion vector candidate.In a reality
It applies in example, when the quantity that may include motion vector candidates in motion vector candidates list is that (here, N is being greater than 0 just to N
Integer) when, motion vector candidates list can be configured to centainly include at least one motion vector candidates.Although when deriving space
It is candidate that at most N number of spatial motion vectors different from each other can be derived when motion vector candidates, but can be determined by any similitude
At least one of N number of spatial motion vectors candidate is removed from motion vector candidates list.Therefore, temporal motion vector is candidate
It can be included in motion vector candidates list.Here, have even if any similitude determination can be indicated when spatial motion vectors
Value different from each other but when the little difference between motion vector, by using maximum value, minimum value, average value, intermediate value or appoint
The method that at least two spaces motion vector set is combined into single spatial motion vectors by meaning weighted sum.It can be by using any similar
Property determine reduce spatial motion vectors candidate quantity.
It selectively, include in motion vector candidates list according to predetermined priority as N number of spatial motion vectors candidate
When, can according to predetermined priority reverse order from motion vector candidates list remove spatial motion vectors candidate at least
One.In other words, can since motion vector candidates list according to sequence from back to front remove spatial motion vectors wait
At least one of choosing.Therefore, temporal motion vector candidate can be included in motion vector candidates list.
Whether can be used for determining according to temporal motion vector candidate whether from above-mentioned motion vector candidates list
It is candidate to remove spatial motion vectors.In addition, according to the quantity of the temporal motion vector candidate for current block or can be used for
The quantity of the temporal motion vector candidate of current block determines that the spatial motion vectors that remove from motion vector candidates list are waited
The quantity of choosing.
Furthermore, it may include (in other words, the quantity of the motion vector candidates in motion vector candidates list can increase by 1
Increase to N+1), so that temporal motion vector candidate is included in motion vector candidates list.
Then, it in step S2004, can be included within any one in the motion vector candidates in motion vector candidates list
It is a to be determined as motion vector predictor.In one embodiment, decoder can be included in motion vector candidates list based on specified
In motion vector candidates in the information of any one (for example, MVP list index) determine the motion-vector prediction of current block
Value.
In step S2005, when the motion vector predictor of current block is determined, can be obtained by using motion vector difference
The motion vector of current block.Motion vector difference can indicate between the motion vector of current block and the motion vector predictor of current block
Difference.The motion vector difference of current block can be coded by entropy/decode.
According to an embodiment of the invention, in order to reduce the information content of motion vector difference, it can be by using adjacent with current block
And the motion vector difference of current block is encoded by the motion vector difference of inter-prediction reconstructed block encoded.At one
In embodiment, the second motion vector difference of current block can be encoded, wherein the second motion vector difference indicates following two fortune
Difference between dynamic vector difference: the motion vector difference and table of the difference between the motion vector of current block and motion-vector prediction are indicated
Show motion vector difference adjacent with current block and by the difference between inter-prediction reconstructed block encoded.
Figure 23 is to show the diagram for deriving the second motion vector difference.
Assuming that the motion vector difference (MVD) of current block (block 2) is (5,5).It here, can be by using on the upside of current block
The motion vector difference of upper block (block 1) the second motion vector difference of current block encoded.
In one embodiment, when assuming that the motion vector difference of upper block is (5,5), due to the motion vector of current block
Difference is identical with the motion vector difference of upper block, therefore the second motion vector difference of current block can be changed to (0,0).Work as motion vector difference
When (0,0) rather than motion vector difference (5,5) are encoded, the letter encoded for the motion vector difference to current block can be reduced
Breath amount.
In addition, when there is the block with motion vector difference identical with the motion vector difference of current block, it can be from contiguous block
Derive motion vector difference of the motion vector difference of current block without sending current block.
Such as above-mentioned example, the second motion vector candidates for deriving current block can be explicitly sent by bit stream
The position of contiguous block or instruction have the position of the contiguous block of motion vector difference identical with the motion vector difference of current block
Information.In one embodiment, it can will be used to identify in the contiguous block of current block by bit stream and be used to derive the second movement
The letter of the contiguous block of vectors candidates or the contiguous block with motion vector candidates identical with the motion vector candidates of current block
Breath (for example, MVD index) is sent to decoder.
In another embodiment, it can implicitly be derived in encoder/decoder according to identical processing and be worked as deriving
The position of the contiguous block of preceding piece of the second motion vector candidates indicates there is fortune identical with the motion vector difference of current block
The information of the position of the contiguous block of dynamic vector difference.In one embodiment, the motion vector predictor as current block (MVP)
Contiguous block motion vector difference be used as the second motion vector difference for deriving current block motion vector difference prediction because
Sub (MVD predictive factor).
When current block by it is bi-directional predicted encoded when, indicate reference picture list 0 (List 0) and reference picture list 1
The whether identical information of the motion vector difference of (List 1) can be encoded.Here, identical motion vector difference can indicate movement arrow
The symbol for measuring difference is identical with size, or can indicate that the size of motion vector difference is identical but the symbol of motion vector difference is different.When
When reference picture list 0 is identical with the motion vector difference of reference picture list 1, reference picture list 0 and reference picture can be omitted
The coding/decoding of any one in the motion vector difference of list 1.
According to another embodiment of the present invention, all pieces in present encoding/decoded picture or the signaling block of band
There can be at least one identical motion vector predictor (MVP) to derive optimal motion vector difference (MVD).Selectively,
According to another embodiment of the present invention, all pieces in present encoding/decoded picture or the signaling block of band can have to
A few identical motion vector difference predicted value (MVD predictive factor) is to derive the second optimal motion vector difference.It here, can needle
Motion vector predictor or motion vector difference predicted value are sent to each signaling block, or can be by using adjacent with signaling block
The encoded information of contiguous block implicitly derives motion vector predictor or motion vector difference predicted value.Here, signaling block can indicate have
There is the area of the size less than at least one resolution ratio in the horizontal resolution and vertical resolution of current picture or current band
Domain.In other words, signaling block can be defined as being less than the presumptive area of the size of current picture or current band.
The information of signaling block can be sent by least one of sequence units, picture unit and slice header.In a reality
Apply in example, can be sent by least one of sequence parameter set, parameter sets and slice header the size of signaling block, form or
At least one of position.Selectively, it can implicitly be pushed away by the encoded information of current block or the contiguous block adjacent with current block
Lead the information of signaling block.Signaling block can have square or rectangular in form, but not limited to this.
Above interframe encode/decoding process can be executed for each luminance signal and carrier chrominance signal.For example, the above interframe is compiled
The acquisition inter-prediction indicator of code/decoding process generates motion vector candidates list, derives motion vector and executes movement and mends
At least one method in repaying can be differently applied to luminance signal and carrier chrominance signal.
Above interframe encode/decoding process can be comparably executed for luminance signal and carrier chrominance signal.For example, execute with
Inter-prediction indicator, motion vector candidates list, the movement arrow of luminance signal are applied to when upper interframe encode/decoding process
At least one of amount candidate, motion vector and reference picture can be equally applied to carrier chrominance signal.
Above method can be executed in the encoder and the decoder by the same way.For example, above interframe encode/decoding
The derivation motion vector candidates list of reason derives motion vector candidates, derives motion vector and execute in motion compensation at least
One method can be equally applicable encoder and decoder.Selectively, the sequence of above method can be applied differently
In encoder and decoder.
Above embodiments of the invention can according to the size of at least one of encoding block, prediction block, block and unit and by
Using.Here, the size can be defined as minimum dimension that above embodiments are applied to or full-size or minimum ruler
Both very little and full-size, or the fixed dimension that embodiment is applied to can be defined as.In addition, in the embodiment above,
First embodiment can be applied to first size, and second embodiment can be applied to the second size.In other words, embodiment can root
The application that is combined according to the size.In addition, above embodiments of the invention can only using minimum dimension or larger size and
Full-size or smaller szie.In other words, above embodiments can be applied to include block size within a predetermined range.
For example, above embodiments can be applied when the size of coding/decoding object block is 8 × 8 or bigger.For example, working as
When the size of coding/decoding object block is 16 × 16 or bigger, above embodiments can be applied.For example, working as coding/decoding target
When the size of block is 32 × 32 or bigger, above embodiments can be applied.For example, when the size of coding/decoding object block is 64
When × 64 or bigger, above embodiments can be applied.For example, when the size of coding/decoding object block is 128 × 128 or bigger
When, above embodiments can be applied.For example, above embodiments can be answered when the size of coding/decoding object block is 4 × 4
With.For example, the size when coding/decoding object block is 8 × 8 or more hour, above embodiments can be applied.For example, ought encode/
The size for decoding object block is 16 × 16 or more hour, and above embodiments can be applied.For example, when coding/decoding object block
It having a size of 8 × 8 or bigger and is 16 × 16 or more hour, above embodiments can be applied.For example, working as coding/decoding target
The size of block is 16 × 16 or bigger and be 64 × 64 or more hour, and above embodiments can be applied.
Above embodiments of the invention can be applied according to time horizon.It can be applied to for identifying above embodiments
The other additional identifiers available signal of time horizon is sent, and above embodiments can be applied to be indicated by corresponding identifier when
Interbed.Here, the identifier can be defined as indicating the smallest tier that can be applied to of embodiment or maximum layer or smallest tier
Both with maximum layer, or can be defined as indicating the certain layer that above embodiments can be applied to.
For example, above embodiments can be applied when the time horizon of current picture is lowermost layer.For example, working as current picture
Time layer identifier be 0 when, above embodiments can be applied.For example, when the time layer identifier of current picture is 1, with
Upper embodiment can be applied.For example, above embodiments can be applied when the time horizon of current picture is top.
Such as above embodiments of the invention, the reference used when generating reference picture list and modifying reference picture list
At least one of reference picture list L0, L1, L2 and L3 can be used in picture collection.
According to an embodiment of the invention, coding/decoding target can be used when calculating boundary intensity in de-blocking filter
At least one of block is to most N number of motion vectors.Here, N is equal to or greater than 1 positive integer, and can be 2,3,4 etc..
When in predicted motion vector motion vector have following unit at least one when, can using it is of the invention with
Upper embodiment: 16- pixel (16-pel) unit, 8- pixel (8-pel) unit, 4- pixel (4-pel) unit, integer-pixel are (whole
Number-pel) unit, 1/2- pixel (1/2-pel) unit, 1/4- pixel (1/4-pel) unit, 1/8- pixel (1/8-pel) list
Position, 1/16- pixel (1/16-pel) unit, 1/32- pixel (1/32-pel) unit and 1/64- pixel (1/64-pel) are single
Position.In addition, can optionally use motion vector according to the above pixel unit when predicted motion vector.
It can define the type of strip of application above embodiments of the invention, and can be according to the application of corresponding type of strip originally
The above embodiments of invention.
For example, prediction block can be by using at least three motion vectors when type of strip is T (three-dimensional prediction)-band
It generates, so as to make the weighted sum be used as coding/decoding object block by the weighted sum for calculating at least three prediction blocks
Final prediction block.For example, prediction block can be sweared by using at least four movements when type of strip is Q (four-way prediction)-band
Amount is to generate, so as to make the weighted sum be used as coding/decoding object block by the weighted sum for calculating at least four prediction blocks
Final prediction block.
Above embodiments of the invention can be applied to inter-prediction and motion compensation process using motion-vector prediction,
And inter-prediction and motion compensation process using skip mode or merging patterns can be applied to.
Can have square form or non-square form using the block form of above embodiments of the invention.
In the above-described embodiments, the method, but this hair are described based on the flow chart with a series of steps or units
The bright sequence for being not limited to the step.But some steps can be performed simultaneously with other steps, or can be pressed with other steps
It is performed according to different order.It will be appreciated by one of ordinary skill in the art that the step in flow chart does not repel one another, and
And in the case where not influencing the scope of the present invention, other steps can be added in flow chart or some steps can be from stream
Journey figure is deleted.
The embodiment being described above includes the example of various aspects.It is, of course, not possible to for description various aspects
Purpose describes the combination that each of component or method are contemplated that, but those of ordinary skill in the art will recognize that further
Combination and permutation are possible.Therefore, this specification be intended to include fall within the spirit and scope of the appended claims it is all
Such alternative forms, modification and change.
Computer readable storage medium may include individual program instruction, data file, data structure etc. or program
The combination of instruction, data file, data structure etc..The program instruction of record in a computer-readable storage medium can especially be set
It counts and is configured to the present invention or be that well known any program refers to for the technical staff of computer software technical field
It enables.The example of computer readable storage medium includes: magnetic recording media (such as hard disk, floppy disk and tape)；Optical storage of data
Medium (such as CD-ROM or DVD-ROM)；Magnet-optical medium (such as floptical)；And it is specially constructed for storing and implementing journey
The hardware device (such as read-only memory (ROM), random access memory (RAM) and flash memory) of sequence instruction.Program instruction shows
Example not only includes the machine language code dissolved by compiler format, further includes that can use the advanced of interpreter implementation by computer
Language codes.Hardware device can be configured to be operated by one or more software modules to carry out processing according to the present invention,
Vice versa.
Although describing the present invention according to specific term (such as detailed elements) and limited embodiments and attached drawing, it
Be only provided to help more generically understand the present invention, present invention is not limited to the above embodiments.It is of the art
The skilled person will understand that various modifications and change can be made from the above description.
Therefore, spirit of the invention should not be so limited to above-described embodiment, whole models of the following claims and their equivalents
Enclosing will be considered within the scope and spirit of the invention.
Industrial applicability
The present invention can be used for the equipment encoded/decoded to image.
Claims (14)
1. a kind of image encoding method, comprising:
Generate the prediction signal of current block；
The residual signals of current block are generated based on prediction signal；
Determine the conversion scheme for being converted to residual signals；And
Quantization is executed to residual signals,
Wherein, the transformation includes transformation and quadratic transformation for the first time, and for the first time in conversion scheme and quadratic transformation scheme extremely
Few one is derived from the coding and rebuilding block adjacent with current block.
2. image encoding method as described in claim 1, wherein prediction signal is generated by intra prediction, and for the first time
At least one of conversion scheme and quadratic transformation scheme are from intra prediction identical with the intra prediction mode of current block
The contiguous block of mode is derived.
3. image encoding method as claimed in claim 2, wherein when with frame identical with the intra prediction mode of current block
When the conversion scheme for the first time of the contiguous block of inner estimation mode indicates that transformation is skipped, the conversion scheme for the first time and quadratic transformation of current block
Scheme is confirmed as transformation and skips.
4. image encoding method as described in claim 1, wherein the quadratic transformation scheme of the residual signals of quantization and scanning are suitable
At least one of sequence is derived from the contiguous block identical with the conversion scheme for the first time of current block of conversion scheme for the first time.
5. image encoding method as described in claim 1, wherein prediction signal is generated by inter-prediction, and for the first time
At least one of conversion scheme and quadratic transformation scheme from motion information identical with the motion information of current block neighbour
Nearly block is derived.
6. image encoding method as claimed in claim 5, wherein motion information include motion vector, reference picture indices and
At least one of reference picture direction.
7. a kind of picture decoding method, comprising:
Obtain the residual signals of the quantization of current block；
Inverse quantization is executed to the residual signals of quantization；And
Determine the conversion scheme for residual signals to be carried out with inverse transformation,
Wherein, the inverse transformation includes transformation and quadratic transformation for the first time, and for the first time in conversion scheme and quadratic transformation scheme
At least one is derived from the decoding and rebuilding block adjacent with current block.
8. picture decoding method as claimed in claim 7, wherein when current block is encoded by intra prediction, become for the first time
At least one of scheme and quadratic transformation scheme are changed from intra prediction mould identical with the intra prediction mode of current block
The contiguous block of formula is derived.
9. picture decoding method as claimed in claim 8, wherein when with frame identical with the intra prediction mode of current block
When the conversion scheme for the first time of the contiguous block of inner estimation mode indicates that transformation is skipped, the conversion scheme for the first time and quadratic transformation of current block
Scheme is confirmed as transformation and skips.
10. picture decoding method as claimed in claim 7, wherein quadratic transformation scheme is from the change for the first time with current block
The contiguous block for changing the identical conversion scheme for the first time of scheme is derived.
11. picture decoding method as claimed in claim 7, wherein when current block is encoded by inter-prediction, become for the first time
At least one of scheme and quadratic transformation scheme are changed from neighbouring with motion information identical with the motion information of current block
Block is derived.
12. picture decoding method as claimed in claim 11, wherein motion information includes motion vector, reference picture indices
At least one of with reference picture direction.
13. a kind of image encoding method, comprising:
Determine the motion vector of current block；
Current block is determined based at least one of the spatial motion vectors candidate of current block and temporal motion vector candidate
Motion vector predictor；
Determining indicates that the first motion-vector prediction of the difference between motion vector and motion vector predictor is poor；And
It is pre- to the second motion vector for indicating the difference between the first motion-vector prediction difference and the motion-vector prediction difference of reconstructed block
Error of measurement is encoded, wherein reconstructed block is adjacent with current block.
14. a kind of picture decoding method, comprising:
Derive at least one of spatial motion vectors candidate and the temporal motion vector candidate of current block；
Generate the motion vector candidates list including at least one of spatial motion vectors candidate and temporal motion vector candidate；
The motion vector predictor of current block is obtained by using motion vector candidates list；
Obtain the second motion vector difference of current block；
The first motion vector difference is obtained based on the motion vector difference of the reconstructed block adjacent with current block and the second motion vector difference；
And
The motion vector of current block is obtained based on the first motion vector difference and motion vector predictor.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210817476.9A CN115052143A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202211602572.8A CN115914625A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202210817457.6A CN115052142A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202211602603.XA CN116016910A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR20160098096 | 2016-08-01 | ||
| KR10-2016-0098096 | 2016-08-01 | ||
| PCT/KR2017/007734 WO2018026118A1 (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
Related Child Applications (4)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210817476.9A Division CN115052143A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202210817457.6A Division CN115052142A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202211602572.8A Division CN115914625A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202211602603.XA Division CN116016910A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109644276A true CN109644276A (en) | 2019-04-16 |
| CN109644276B CN109644276B (en) | 2022-12-30 |
Family
ID=61073484
Family Applications (5)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211602572.8A Pending CN115914625A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN201780048129.1A Active CN109644276B (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202211602603.XA Pending CN116016910A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202210817457.6A Pending CN115052142A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202210817476.9A Pending CN115052143A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211602572.8A Pending CN115914625A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
Family Applications After (3)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211602603.XA Pending CN116016910A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202210817457.6A Pending CN115052142A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
| CN202210817476.9A Pending CN115052143A (en) | 2016-08-01 | 2017-07-18 | Image encoding/decoding method |
Country Status (3)
| Country | Link |
|---|---|
| KR (5) | KR102321394B1 (en) |
| CN (5) | CN115914625A (en) |
| WO (1) | WO2018026118A1 (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020211765A1 (en) * | 2019-04-17 | 2020-10-22 | Huawei Technologies Co., Ltd. | An encoder, a decoder and corresponding methods harmonzting matrix-based intra prediction and secoundary transform core selection |
| WO2021110045A1 (en) * | 2019-12-03 | 2021-06-10 | Huawei Technologies Co., Ltd. | Coding method, device, system with merge mode |
| CN113287310A (en) * | 2019-06-10 | 2021-08-20 | 株式会社 Xris | Method for encoding/decoding image signal and apparatus thereof |
| CN113473129A (en) * | 2020-03-30 | 2021-10-01 | 杭州海康威视数字技术股份有限公司 | Encoding and decoding method and device |
| CN113812162A (en) * | 2019-05-10 | 2021-12-17 | 北京字节跳动网络技术有限公司 | Context modeling for simplified quadratic transforms in video |
| CN114073078A (en) * | 2019-06-23 | 2022-02-18 | Lg 电子株式会社 | Method and apparatus for syntax signaling in video/image coding system |
| CN114080802A (en) * | 2019-07-01 | 2022-02-22 | 高通股份有限公司 | Minimum allowed quantization parameter for transform skip mode and palette mode in video coding |
| CN114762341A (en) * | 2019-10-08 | 2022-07-15 | Lg电子株式会社 | Method and apparatus for transform-based image coding |
| US11575901B2 (en) | 2019-08-17 | 2023-02-07 | Beijing Bytedance Network Technology Co., Ltd. | Context modeling of side information for reduced secondary transforms in video |
| US11638008B2 (en) | 2019-08-03 | 2023-04-25 | Beijing Bytedance Network Technology Co., Ltd. | Selection of matrices for reduced secondary transform in video coding |
| US11924469B2 (en) | 2019-06-07 | 2024-03-05 | Beijing Bytedance Network Technology Co., Ltd. | Conditional signaling of reduced secondary transform in video bitstreams |
| US12149680B2 (en) | 2019-12-03 | 2024-11-19 | Huawei Technologies Co., Ltd. | Coding method, device, system with merge mode |
Families Citing this family (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019164031A1 (en) * | 2018-02-22 | 2019-08-29 | 엘지전자 주식회사 | Image decoding method and apparatus according to block division structure in image coding system |
| US20200413040A1 (en) * | 2018-03-14 | 2020-12-31 | Electronics And Telecommunications Research Institute | Method and device for encoding/decoding image and recording medium having bitstream stored thereon |
| CN118381901A (en) * | 2018-03-21 | 2024-07-23 | Lx 半导体科技有限公司 | Image encoding/decoding apparatus and apparatus for transmitting image data |
| WO2019190098A1 (en) * | 2018-03-28 | 2019-10-03 | 한국전자통신연구원 | Method and device for image encoding/decoding, and recording medium for storing bit stream |
| WO2019190282A1 (en) * | 2018-03-29 | 2019-10-03 | 엘지전자 주식회사 | Method and device for processing video signal on basis of multiplication-free rotation-based transform |
| KR102465119B1 (en) | 2018-04-01 | 2022-11-09 | 엘지전자 주식회사 | Method and apparatus for processing a video signal using reduced quadratic transform |
| CN111937386B (en) | 2018-04-01 | 2023-09-01 | Lg电子株式会社 | Method and apparatus for processing video signal by applying secondary transform to partition block |
| KR102516233B1 (en) * | 2018-04-02 | 2023-03-30 | 엘지전자 주식회사 | Image coding method and device based on motion vector |
| JP7395497B2 (en) * | 2018-05-07 | 2023-12-11 | インターデジタル ヴイシー ホールディングス， インコーポレイテッド | Data dependencies in encoding/decoding |
| FI3806476T3 (en) * | 2018-06-03 | 2023-08-15 | Lg Electronics Inc | Method and device for processing video signal by using reduced transform |
| CN112166609B (en) * | 2018-06-08 | 2024-10-18 | 株式会社Kt | Method and apparatus for processing video signal |
| CN116828174A (en) * | 2018-06-27 | 2023-09-29 | Lg电子株式会社 | Method for encoding and decoding video signal and transmitting method |
| CN110868589B (en) * | 2018-08-28 | 2023-10-20 | 华为技术有限公司 | Inter-frame prediction method and device and coding/decoding method and device applied by same |
| BR112021003917A2 (en) | 2018-08-28 | 2021-05-18 | Huawei Technologies Co., Ltd. | method and apparatus for building candidate movement information list, interprediction method, and apparatus |
| CN110876065A (en) * | 2018-08-29 | 2020-03-10 | 华为技术有限公司 | Construction method of candidate motion information list, and inter-frame prediction method and device |
| JP2021509789A (en) | 2018-09-02 | 2021-04-01 | エルジー エレクトロニクス インコーポレイティド | Video signal coding / decoding method and equipment for that purpose |
| CN116055719A (en) * | 2018-09-05 | 2023-05-02 | Lg电子株式会社 | Apparatus for encoding/decoding image signal and apparatus for transmitting image signal |
| CA3111118A1 (en) | 2018-09-21 | 2020-03-26 | Huawei Technologies Co., Ltd. | Apparatus and method for inverse quantization |
| WO2020106653A1 (en) * | 2018-11-20 | 2020-05-28 | Interdigital Vc Holdings, Inc. | Current picture referencing block vector initialization with dual tree |
| KR102494697B1 (en) * | 2018-12-21 | 2023-02-06 | 삼성전자주식회사 | Image encoding apparatus and image decoding apparatus using triangle prediction mode, and image encoding method and image decoding method thereby |
| US11711534B2 (en) * | 2019-02-28 | 2023-07-25 | Apple Inc. | Method for encoding/decoding image signal, and device therefor |
| CN113574880B (en) * | 2019-03-13 | 2023-04-07 | 北京字节跳动网络技术有限公司 | Partitioning with respect to sub-block transform modes |
| CN113994672B (en) * | 2019-04-25 | 2023-07-25 | 北京达佳互联信息技术有限公司 | Method and apparatus for video encoding and decoding using triangle prediction |
| CN113170130A (en) * | 2019-05-02 | 2021-07-23 | 株式会社 Xris | Image signal encoding/decoding method and apparatus thereof |
| JP7256296B2 (en) * | 2019-05-08 | 2023-04-11 | エルジー エレクトロニクス インコーポレイティド | Image encoding/decoding method, apparatus and method for transmitting bitstream with MIP and LFNST |
| KR20220024500A (en) | 2019-07-12 | 2022-03-03 | 엘지전자 주식회사 | Transformation-based video coding method and apparatus |
| CN116405697B (en) * | 2019-07-23 | 2024-10-29 | 杭州海康威视数字技术股份有限公司 | Encoding and decoding method, device and equipment thereof |
| JP7267885B2 (en) * | 2019-09-20 | 2023-05-02 | Kddi株式会社 | Image decoding device, image decoding method and program |
| CN115606182A (en) * | 2020-03-25 | 2023-01-13 | 抖音视界有限公司（Cn） | Codec video processing using enhanced quadratic transform |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110115986A (en) * | 2010-04-16 | 2011-10-24 | 에스케이 텔레콤주식회사 | Video coding and decoding method and apparatus |
| CN103650504A (en) * | 2011-05-24 | 2014-03-19 | 高通股份有限公司 | Control of video encoding based on image capture parameters |
| KR20140053155A (en) * | 2011-07-01 | 2014-05-07 | 삼성전자주식회사 | Mode-dependent transforms for residual coding with low latency |
| CN103780911A (en) * | 2011-06-20 | 2014-05-07 | 韩国电子通信研究院 | Video decoding apparatus |
| CN104159108A (en) * | 2014-08-11 | 2014-11-19 | 浙江大学 | Real-time lossless electrocardiosignal compression method and device based on adaptive prediction and variable length coding improvement |
| US20150208073A1 (en) * | 2003-07-18 | 2015-07-23 | Samsung Electronics Co., Ltd. | Image encoding and decoding apparatus and method |
| CN104980755A (en) * | 2010-01-14 | 2015-10-14 | 三星电子株式会社 | Method And Apparatus For Encoding And Decoding Video By Using Pattern Information In Hierarchical Data Unit |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080107965A (en) * | 2007-06-08 | 2008-12-11 | 삼성전자주식회사 | Method and apparatus for encoding and decoding image using object boundary based partition |
| KR20100071865A (en) * | 2008-12-19 | 2010-06-29 | 삼성전자주식회사 | Method for constructing and decoding a video frame in a video signal processing apparatus using multi-core processor and apparatus thereof |
| EP2547108A4 (en) * | 2010-04-13 | 2014-07-02 | Samsung Electronics Co Ltd | Video-encoding method and video-encoding apparatus using prediction units based on encoding units determined in accordance with a tree structure, and video-decoding method and video-decoding apparatus using prediction units based on encoding units determined in accordance with a tree structure |
| CN106231338B (en) * | 2010-04-13 | 2019-08-27 | Ge视频压缩有限责任公司 | Decoder, coding/decoding method, encoder and coding method |
| ES2699073T3 (en) * | 2010-08-17 | 2019-02-07 | Samsung Electronics Co Ltd | Video decoding procedure using variable tree structure transformation unit |
| CN103081470B (en) * | 2010-09-02 | 2016-08-03 | Lg电子株式会社 | The method of coding and decoding video and the device using the method |
| WO2013023005A1 (en) * | 2011-08-08 | 2013-02-14 | General Instrument Corporation | Residual tree structure of transform unit partitioning |
| CN107959857B (en) * | 2011-10-18 | 2022-03-01 | 株式会社Kt | Video signal decoding method |
| GB2509881B (en) * | 2011-11-08 | 2018-09-12 | Kt Corp | A method of decoding a video signal using an AMVP mode |
| FR2993084A1 (en) * | 2012-07-09 | 2014-01-10 | France Telecom | VIDEO CODING METHOD BY PREDICTING CURRENT BLOCK PARTITIONING, DECODING METHOD, CODING AND DECODING DEVICES AND CORRESPONDING COMPUTER PROGRAMS |
| KR20140008503A (en) * | 2012-07-10 | 2014-01-21 | 한국전자통신연구원 | Method and apparatus for image encoding/decoding |
| KR20140043015A (en) * | 2012-09-28 | 2014-04-08 | 연세대학교 산학협력단 | Method and apparatus for image encoding |
| CN110545422B (en) * | 2012-10-12 | 2022-11-22 | 韩国电子通信研究院 | Image encoding/decoding method and apparatus using the same |
| EP2952003B1 (en) * | 2013-01-30 | 2019-07-17 | Intel Corporation | Content adaptive partitioning for prediction and coding for next generation video |
| KR20160045243A (en) * | 2014-10-17 | 2016-04-27 | 한국항공대학교산학협력단 | Video coding Apparatus of Planar intra prediction and method thereof |
-
2017
- 2017-07-18 CN CN202211602572.8A patent/CN115914625A/en active Pending
- 2017-07-18 CN CN201780048129.1A patent/CN109644276B/en active Active
- 2017-07-18 CN CN202211602603.XA patent/CN116016910A/en active Pending
- 2017-07-18 KR KR1020170091138A patent/KR102321394B1/en active IP Right Grant
- 2017-07-18 CN CN202210817457.6A patent/CN115052142A/en active Pending
- 2017-07-18 WO PCT/KR2017/007734 patent/WO2018026118A1/en active Application Filing
- 2017-07-18 CN CN202210817476.9A patent/CN115052143A/en active Pending
-
2021
- 2021-10-28 KR KR1020210145695A patent/KR102400315B1/en active IP Right Grant
-
2022
- 2022-05-17 KR KR1020220059955A patent/KR102549022B1/en active Application Filing
-
2023
- 2023-06-23 KR KR1020230080874A patent/KR102681156B1/en active IP Right Grant
-
2024
- 2024-06-28 KR KR1020240084944A patent/KR20240110760A/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150208073A1 (en) * | 2003-07-18 | 2015-07-23 | Samsung Electronics Co., Ltd. | Image encoding and decoding apparatus and method |
| CN104980755A (en) * | 2010-01-14 | 2015-10-14 | 三星电子株式会社 | Method And Apparatus For Encoding And Decoding Video By Using Pattern Information In Hierarchical Data Unit |
| KR20110115986A (en) * | 2010-04-16 | 2011-10-24 | 에스케이 텔레콤주식회사 | Video coding and decoding method and apparatus |
| CN103650504A (en) * | 2011-05-24 | 2014-03-19 | 高通股份有限公司 | Control of video encoding based on image capture parameters |
| CN103780911A (en) * | 2011-06-20 | 2014-05-07 | 韩国电子通信研究院 | Video decoding apparatus |
| KR20140053155A (en) * | 2011-07-01 | 2014-05-07 | 삼성전자주식회사 | Mode-dependent transforms for residual coding with low latency |
| CN104159108A (en) * | 2014-08-11 | 2014-11-19 | 浙江大学 | Real-time lossless electrocardiosignal compression method and device based on adaptive prediction and variable length coding improvement |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020211765A1 (en) * | 2019-04-17 | 2020-10-22 | Huawei Technologies Co., Ltd. | An encoder, a decoder and corresponding methods harmonzting matrix-based intra prediction and secoundary transform core selection |
| US11611779B2 (en) | 2019-05-10 | 2023-03-21 | Beijing Bytedance Network Technology Co., Ltd. | Multiple secondary transform matrices for video processing |
| CN113812162A (en) * | 2019-05-10 | 2021-12-17 | 北京字节跳动网络技术有限公司 | Context modeling for simplified quadratic transforms in video |
| US11622131B2 (en) | 2019-05-10 | 2023-04-04 | Beijing Bytedance Network Technology Co., Ltd. | Luma based secondary transform matrix selection for video processing |
| CN113812162B (en) * | 2019-05-10 | 2022-12-30 | 北京字节跳动网络技术有限公司 | Context modeling for simplified quadratic transforms in video |
| US11575940B2 (en) | 2019-05-10 | 2023-02-07 | Beijing Bytedance Network Technology Co., Ltd. | Context modeling of reduced secondary transforms in video |
| US11924469B2 (en) | 2019-06-07 | 2024-03-05 | Beijing Bytedance Network Technology Co., Ltd. | Conditional signaling of reduced secondary transform in video bitstreams |
| CN113287310A (en) * | 2019-06-10 | 2021-08-20 | 株式会社 Xris | Method for encoding/decoding image signal and apparatus thereof |
| CN114073078A (en) * | 2019-06-23 | 2022-02-18 | Lg 电子株式会社 | Method and apparatus for syntax signaling in video/image coding system |
| CN114080802A (en) * | 2019-07-01 | 2022-02-22 | 高通股份有限公司 | Minimum allowed quantization parameter for transform skip mode and palette mode in video coding |
| US11882274B2 (en) | 2019-08-03 | 2024-01-23 | Beijing Bytedance Network Technology Co., Ltd | Position based mode derivation in reduced secondary transforms for video |
| US11638008B2 (en) | 2019-08-03 | 2023-04-25 | Beijing Bytedance Network Technology Co., Ltd. | Selection of matrices for reduced secondary transform in video coding |
| US11575901B2 (en) | 2019-08-17 | 2023-02-07 | Beijing Bytedance Network Technology Co., Ltd. | Context modeling of side information for reduced secondary transforms in video |
| US11968367B2 (en) | 2019-08-17 | 2024-04-23 | Beijing Bytedance Network Technology Co., Ltd. | Context modeling of side information for reduced secondary transforms in video |
| CN114762341A (en) * | 2019-10-08 | 2022-07-15 | Lg电子株式会社 | Method and apparatus for transform-based image coding |
| CN114762341B (en) * | 2019-10-08 | 2024-01-16 | Lg电子株式会社 | Method and apparatus for transform-based image coding |
| WO2021110045A1 (en) * | 2019-12-03 | 2021-06-10 | Huawei Technologies Co., Ltd. | Coding method, device, system with merge mode |
| US12149680B2 (en) | 2019-12-03 | 2024-11-19 | Huawei Technologies Co., Ltd. | Coding method, device, system with merge mode |
| CN113473129B (en) * | 2020-03-30 | 2022-12-23 | 杭州海康威视数字技术股份有限公司 | Encoding and decoding method and device |
| CN113473129A (en) * | 2020-03-30 | 2021-10-01 | 杭州海康威视数字技术股份有限公司 | Encoding and decoding method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| KR102549022B1 (en) | 2023-06-27 |
| KR20220068974A (en) | 2022-05-26 |
| KR20210133202A (en) | 2021-11-05 |
| KR20180014655A (en) | 2018-02-09 |
| KR20240110760A (en) | 2024-07-16 |
| CN116016910A (en) | 2023-04-25 |
| KR102400315B1 (en) | 2022-05-23 |
| CN115052143A (en) | 2022-09-13 |
| WO2018026118A1 (en) | 2018-02-08 |
| KR102681156B1 (en) | 2024-07-03 |
| CN109644276B (en) | 2022-12-30 |
| CN115052142A (en) | 2022-09-13 |
| KR20230096953A (en) | 2023-06-30 |
| KR102321394B1 (en) | 2021-11-03 |
| CN115914625A (en) | 2023-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109644276A (en) | Image coding/decoding method | |
| CN109804627A (en) | Image coding/decoding method and equipment | |
| CN109196864A (en) | Image coding/decoding method and recording medium for the method | |
| CN110024394A (en) | The recording medium of method and apparatus and stored bits stream to encoding/decoding image | |
| CN109479141A (en) | Image coding/decoding method and recording medium for the method | |
| CN109417617A (en) | Intra-frame prediction method and device | |
| CN109417636A (en) | Method and apparatus for the encoding/decoding image based on transformation | |
| CN109417629A (en) | Image coding/decoding method and recording medium for this method | |
| CN104185988B (en) | Method and apparatus for sharing candidate list | |
| CN109479129A (en) | The recording medium of image coding/decoding method and device and stored bits stream | |
| CN109804626A (en) | Method and apparatus for being coded and decoded to image and the recording medium for stored bits stream | |
| CN110463201A (en) | Use the prediction technique and device of reference block | |
| CN109952762A (en) | The recording medium of video coding/decoding method and equipment and stored bits stream | |
| CN110024399A (en) | The recording medium of method and apparatus and stored bits stream to encoding/decoding image | |
| CN109479138A (en) | Image coding/decoding method and device | |
| CN110121884A (en) | Method and apparatus for filtering | |
| CN109792515A (en) | The recording medium of image coding/decoding method and device and stored bits stream | |
| CN109565593A (en) | The recording medium of image coding/decoding method and equipment and stored bits stream | |
| CN109997363A (en) | Image coding/decoding method and device and the recording medium for being stored with bit stream | |
| CN109644272A (en) | Geometric type priority for construction candidate list | |
| CN107925759A (en) | Method and apparatus for coding and decoding infra-frame prediction | |
| CN110476425A (en) | Prediction technique and device based on block form | |
| CN108353166A (en) | Method and apparatus for encoding/decoding image | |
| CN109691099A (en) | Video coding/decoding method and device and recording medium in wherein stored bits stream | |
| CN109314785A (en) | Method and apparatus for exporting motion prediction information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |