CN102685479A - Video encoding and decoding processing method and device - Google Patents
Video encoding and decoding processing method and device Download PDFInfo
- Publication number
- CN102685479A CN102685479A CN2011100600562A CN201110060056A CN102685479A CN 102685479 A CN102685479 A CN 102685479A CN 2011100600562 A CN2011100600562 A CN 2011100600562A CN 201110060056 A CN201110060056 A CN 201110060056A CN 102685479 A CN102685479 A CN 102685479A
- Authority
- CN
- China
- Prior art keywords
- candidate
- derivation
- moving
- derivation information
- information
- 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.)
- Pending
Links
Images
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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/513—Processing of motion vectors
- H04N19/517—Processing of motion vectors by encoding
-
- 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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/56—Motion estimation with initialisation of the vector search, e.g. estimating a good candidate to initiate a search
Abstract
An embodiment of the invention provides a video encoding and decoding processing method and a device. The encoding processing method includes obtaining a first-level candidate motion derived information set of a current block; if the number of pieces of candidate motion derived information in the first-level candidate motion derived information set is larger than 2, merging the candidate motion derived information in the first-level candidate motion derived information set to obtain a second-level candidate motion derived information set, the number of pieces of candidate motion derived information comprised in the second-level candidate motion derived information set being smaller than or equal to 2; and encoding the current block according to the candidate motion derived information comprised in the second-level candidate motion derived information set. According to the embodiment of the video encoding and decoding processing method and the device, the problem that an error occurs in the process of decoding an encoded code stream of the current block caused by the incorrectly estimated number of candidate motion derived information in the prior art can be solved.
Description
Technical field
The embodiment of the invention relates to the image/video technology, relates in particular to a kind of coding and decoding video processing method and device.
Background technology
Along with improving constantly of network technology and terminal processing capacity; People are to present widely used series 2 (the Moving Pictures Experts Group-2 of Motion Picture Experts Group; Hereinafter to be referred as: MPEG-2); Motion Picture Experts Group's series 4 (hereinafter to be referred as: MPEG-4), H.264 wait video coding and decoding technology to propose new requirement, for example hope to satisfy the requirement of high definition, 3D; Thereby satisfy the application of frontiers such as new home theater, remote monitoring, digital broadcasting, mobile flow medium, mobile imaging, medical imaging; In order to satisfy these application, video encoding standard of new generation (High Efficiency Video Coding, hereinafter to be referred as: HEVC) arise at the historic moment.
Existing video coding often adopts inter prediction to eliminate the time domain redundancy, thereby video is compressed.In inter prediction encoding, need searching motion vector (Motion Vector, hereinafter to be referred as: MV) confirm present encoding piece the most similar reference block in reference frame, and this reference block and original video compared, calculate residual error.In order further to improve compression ratio; Also increased motion-vector prediction (MotionVector Prediction; Hereinafter to be referred as: MVP) module; Predict the MV of current block through the MV of the adjacent block of having encoded, and with motion vector difference (Motion Vector Difference, hereinafter to be referred as: MVD) represent poor between MV and the MVP.Finally, inter prediction encoding carries MVD and residual error in code stream, thereby greatly reduces data volume.HEVC is in order to improve compression efficiency, and deriving at the MV that utilizes encoding block has increased many improvement technology aspect the MV of current block, comprising to the improvement of MVP module, to the improvement of MVP index coding and increased merging (hereinafter to be referred as Merge) pattern.Specifically, prior art can select for use on the adjacent MVP in a spatial domain left side, the spatial domain three of adjacent MVP and time domain MVP as the candidate in the MVP module respectively.Fig. 1 is the searching method sketch map of spatial domain MVP in the prior art, and is as shown in Figure 1, and the piece that (F, G, H, I) selects first MV to use from top to bottom in all adjacent pieces of left side is with the MV of this piece adjacent MVP in a left side, spatial domain as current block; The piece that (A, B, C, D, E) selects first MV to use from right to left in all adjacent up pieces is with adjacent MVP on the spatial domain of MV as current block of this piece.Fig. 2 is the searching method sketch map of time domain MVP in the prior art, and is as shown in Figure 2, in reference frame, selects the reference block with the current block correspondence position, with the MV of this reference block center time domain MVP as current block.Through above-mentioned searching method, can obtain three candidate MVP, from three candidate MVP, select optimum MVP according to rate distortion costs, and the index of this MVP in candidate MVP write code stream.When the MVP index was encoded, prior art was in order to reduce the bit number of use, instructed the coding of MVP index according to the number of candidate MVP, particularly according to MVP number judge whether to encode MVP index and infer the shared bit number of MVP index.In addition; HEVC has also increased the Merge pattern in the process that movable information is derived; Promptly get the candidate MV of the MV of the adjacent encoding block of spatial domain and time domain, and among the MV that candidate MV that all Merge patterns are chosen and motion search go out, choose the MV of the superior as current block according to rate distortion costs as current block.If the MV of current block is the MV that chooses in the Merge pattern, when coding, only need indicates this piece so and use the Merge pattern and indicated the index of the used MV of this piece in candidate MV.In order to reduce coding MV index required bit number, prior art instructs the coding of MV index according to the number of candidate MV, particularly according to candidate MV number judge whether to encode MV index and infer the shared bit number of MV index.No matter be candidate MVP or the candidate is MV, it all can be called candidate's derivation information of moving, and the set of being made up of candidate MVP or candidate MV then can be called candidate's derivation ensemble of communication of moving.
Yet; Coding and the resolving of derivation information depend on the move number of derivation information of candidate because the candidate moves; In case generation packet loss; The time domain candidate of decoding end moves derivation information possibly can't be searched, so that the move number of derivation information of the candidate who estimates is incorrect, thereby cause the process that the encoding code stream of current block is decoded is made mistakes.
Summary of the invention
The embodiment of the invention provides a kind of coding and decoding video processing method and device, and is incorrect because of the move number of derivation information of the candidate who estimates to solve prior art, and the problem that the process that the encoding code stream to current block that causes is decoded is made mistakes.
The embodiment of the invention provides a kind of video coding processing method, comprising:
The first order candidate who the obtains current block derivation ensemble of communication of moving;
Candidate in the derivation ensemble of communication moves the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
According to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out encoding process.
The embodiment of the invention provides another kind of video coding processing method, comprising:
The candidate who the obtains current block derivation ensemble of communication of moving;
There is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, then said current block carried out encoding process, obtain the encoding code stream of said current block, do not comprise motion derivation information index coding in the said encoding code stream.
The embodiment of the invention provides a kind of video decode processing method, comprising:
Receive the encoding code stream of current block;
The first order candidate who the obtains current block derivation ensemble of communication of moving;
Candidate in the derivation ensemble of communication moves the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
According to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving the encoding code stream of said current block is carried out decoding processing.
The embodiment of the invention provides another kind of video decode processing method, comprising:
Receive the encoding code stream of current block;
The candidate who the obtains current block derivation ensemble of communication of moving;
There is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, then the encoding code stream of said current block carried out decoding processing according to the non-spatial domain candidate derivation information of moving.
The embodiment of the invention provides a kind of code device, comprising:
First acquisition module, the first order candidate who is used to the to obtain current block derivation ensemble of communication of moving;
First merges processing module; Be used for the candidate of derivation ensemble of communication and move the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
The first encoding process module is used for candidate that the derivation ensemble of communication the comprises derivation information of moving of moving according to said second level candidate, and said current block is carried out encoding process.
The embodiment of the invention provides another kind of code device, comprising:
Second acquisition module, the candidate who is used to the to obtain current block derivation ensemble of communication of moving;
The second encoding process module; Be used for that there is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves; Then said current block is carried out encoding process, obtain the encoding code stream of said current block, do not comprise motion derivation information index coding in the said encoding code stream.
The embodiment of the invention provides a kind of decoding device, comprising:
First receiver module is used to receive the encoding code stream of current block;
The 3rd acquisition module, the first order candidate who is used to the to obtain said current block derivation ensemble of communication of moving;
Second merges processing module; Be used for the candidate of derivation ensemble of communication and move the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
First decoding process module carries out decoding processing according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving to the encoding code stream of said current block.
The embodiment of the invention provides another kind of decoding device, comprising:
Second receiver module is used to receive the encoding code stream of current block;
The 4th acquisition module, the candidate who is used to the to obtain current block derivation ensemble of communication of moving;
Second decoding process module is used for that there is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, and then according to the non-spatial domain candidate derivation information of moving the encoding code stream of said current block is carried out decoding processing.
In the embodiment of the invention; Coding side can move the number of derivation information the candidate during first order candidate moves the derivation ensemble of communication greater than under 2 the situation; First order candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving is processed at the most two candidates derivation information of moving through merging; To obtain the second level candidate derivation ensemble of communication of moving; And use this second level candidate derivation ensemble of communication of moving and carry out encoding process, make this second level candidate is moved the motion derivation information index selected in the derivation ensemble of communication when encoding, need a bit at the most.Decoding end can be moved after the derivation ensemble of communication adopting above-mentioned identical merging processing mode to obtain second level candidate; No matter the situation that whether exists reference frame to lose; Decoding end only need be carried out decoding processing to the index coding of a bit; And can not occur thinking the part of index coding other part of the encoding code stream that is current block by mistake, thereby can avoid the encoding code stream of current block is decoded by error.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art; To do one to the accompanying drawing of required use in embodiment or the description of the Prior Art below introduces simply; Obviously, the accompanying drawing in describing below is some embodiments of the present invention, for those of ordinary skills; Under the prerequisite of not paying creative work property, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the searching method sketch map of spatial domain MVP in the prior art;
Fig. 2 is the searching method sketch map of time domain MVP in the prior art;
Fig. 3 is the flow chart of video coding processing method embodiment one of the present invention;
Fig. 4 is the flow chart of video coding processing method embodiment two of the present invention;
Fig. 5 is the flow chart of video coding processing method embodiment three of the present invention;
Fig. 6 is the flow chart of video coding processing method embodiment four of the present invention;
Fig. 7 is the flow chart of video coding processing method embodiment five of the present invention;
Fig. 8 is the flow chart of video coding processing method embodiment six of the present invention;
Fig. 9 is the flow chart of video coding processing method embodiment seven of the present invention;
Figure 10 is the flow chart of video coding processing method embodiment eight of the present invention;
Figure 11 is the flow chart of video coding processing method embodiment nine of the present invention;
Figure 12 is the flow chart of video coding processing method embodiment ten of the present invention;
Figure 13 is the flow chart of video decode processing method embodiment one of the present invention;
Figure 14 is the flow chart of video decode processing method embodiment two of the present invention;
Figure 15 is the flow chart of video decode processing method embodiment three of the present invention;
Figure 16 is the flow chart of video decode processing method embodiment four of the present invention;
Figure 17 is the flow chart of video decode processing method embodiment five of the present invention;
Figure 18 is the flow chart of video decode processing method embodiment six of the present invention;
Figure 19 is the flow chart of video decode processing method embodiment seven of the present invention;
Figure 20 is the flow chart of video decode processing method embodiment eight of the present invention;
Figure 21 is the flow chart of video decode processing method embodiment nine of the present invention;
Figure 22 is the flow chart of video decode processing method embodiment ten of the present invention;
Figure 23 is the structural representation of code device embodiment one of the present invention;
Figure 24 is the structural representation of code device embodiment two of the present invention;
Figure 25 is the structural representation of decoding device embodiment one of the present invention;
Figure 26 is the structural representation of decoding device embodiment two of the present invention.
Embodiment
For the purpose, technical scheme and the advantage that make the embodiment of the invention clearer; To combine the accompanying drawing in the embodiment of the invention below; Technical scheme in the embodiment of the invention is carried out clear, intactly description; Obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills are not making the every other embodiment that is obtained under the creative work prerequisite, all belong to the scope of the present invention's protection.
Fig. 3 is the flow chart of video coding processing method embodiment one of the present invention, and as shown in Figure 3, the method for present embodiment can comprise:
For instance, the code device first order candidate that can the obtain current block derivation ensemble of communication of moving.So-called current block is current video blocks to be encoded.Motion derivation information in the present embodiment can be the MV under MVP or the Merge pattern.It will be appreciated by persons skilled in the art that the technical scheme that is used for other motion derivation information that video blocks is carried out inter prediction is also gone for present embodiment.
With motion derivation information be MVP for instance; Step 301 in the present embodiment can adopt existing techniques in realizing; Piece for example shown in Figure 1, that (F, G, H, I) selects first MV to use from top to bottom in all adjacent pieces of left side is with the MV of this piece adjacent MVP in a left side, spatial domain as current block; The piece that (A, B, C, D, E) selects first MV to use from right to left in all adjacent up pieces is with adjacent MVP on the spatial domain of MV as current block of this piece; As shown in Figure 2, in reference frame, select the reference block with the current block correspondence position, with the MV of this reference block center time domain MVP as current block.Owing to any MVP among the adjacent MVP on above-mentioned time domain MVP, the adjacent MVP in a left side, spatial domain, the spatial domain might not exist; Therefore, adopt the first order candidate MVP set of the method acquisition of prior art can comprise following any set form: { time domain MVP}, { the adjacent MVP} in a left side, spatial domain, { adjacent MVP}, { time domain MVP, the adjacent MVP} in a left side, spatial domain, { adjacent MVP}, { adjacent MVP} and { adjacent MVP} on time domain MVP, the left adjacent MVP in spatial domain, the spatial domain on the adjacent MVP in a left side, spatial domain, the spatial domain on time domain MVP, the spatial domain on the spatial domain.
With motion derivation information be under the Merge pattern MV for instance, the step 301 in the present embodiment also can adopt existing techniques in realizing.Candidate MV possibly comprise on the adjacent MV in a spatial domain left side, the spatial domain adjacent MV under adjacent MV, time domain MV, the upper right adjacent MV in spatial domain, the left side, spatial domain; Therefore; With above-mentioned candidate MVP similarly, first order candidate MV set also can include some kinds of set forms of at least one composition among above-mentioned five kinds of candidate MV.
It will be appreciated by persons skilled in the art that step 301 also can not adopt existing techniques in realizing, for example, for candidate MVP, other time domain candidate MVP more than other spatial domain candidate MVP perhaps comprised more more than it also possibly comprise more; For the candidate MV under the Merge pattern, other time domain candidate MV more than other spatial domain candidate MV perhaps comprised more more than it also possibly comprise more.The first order candidate that present embodiment does not obtain code device the move particular content of derivation information of the candidate that comprises in the derivation ensemble of communication that moves limits, and gets final product so long as can carry out the motion derivation information of inter prediction encoding.
Table 1 is for being directed against a kind of coded system of three candidate MVP that possibly exist in the prior art.
Table 1
| MVPIdx | MVPNum＝1 | ?MVPNum＝2 | ?MVPNum＝3 |
| 0 | (inferred) | 0 | 0 |
| 1 | 1 | 10 | |
| 2 | 11 |
As shown in table 1, wherein MVPIdx is the index of MVP in first order candidate MVP set of selection, and MVPNum is the quantity of the candidate MVP in the first order candidate MVP set.Specifically; If only comprise 1 candidate MVP in the first order candidate MVP set of adopting mode illustrated in figures 1 and 2 to obtain; It is the situation of corresponding MVPNum=1; Then the index of this candidate MVP is not encoded, this unique candidate MVP possibly be any among the adjacent candidate MVP on time domain candidate MVP, the spatial domain adjacent candidate MVP in a left side and the spatial domain; If comprise 2 candidate MVP in the first order candidate MVP set of adopting mode illustrated in figures 1 and 2 to obtain; It is the situation of corresponding MVPNum=2; The corresponding MVPIdx of the 1st candidate MVP in the first order candidate MVP set is 0, and its corresponding index is encoded to 0, and the corresponding MVPIdx of the 2nd candidate MVP in the first order candidate MVP set is 1; Its corresponding index is encoded to 1, and these two candidate MVP possibly be any two among the adjacent candidate MVP on time domain candidate MVP, the spatial domain adjacent candidate MVP in a left side and the spatial domain; If comprise 3 candidate MVP in the first order candidate MVP set of adopting mode illustrated in figures 1 and 2 to obtain; It is the situation of corresponding MVPNum=3; The corresponding MVPIdx of the 1st candidate MVP in the first order candidate MVP set is 0, and its corresponding index is encoded to 0, and the corresponding MVPIdx of the 2nd candidate MVP in the first order candidate MVP set is 1; Its corresponding index is encoded to 10; The corresponding MVPIdx of the 3rd candidate MVP in the first order candidate MVP set is 2, and its corresponding index is encoded to 11, and these 3 candidate MVP are adjacent candidate MVP on time domain candidate MVP, the spatial domain adjacent candidate MVP in a left side and the spatial domain.Hence one can see that, and the coded system of prior art is relevant with the number of candidate MVP, because above-mentioned possible candidate MVP has 3; Therefore, when prior art was encoded to the index of these 3 candidate MVP in first order candidate MVP set, this index coding possibly be a bit; Also possibly be two bits, therefore, decoding end is when carrying out decoding processing to the index coding; Both possibly need bit of decoding, two bits of also need decoding, the quantity of candidate MVP during this first order candidate MVP that depends on that decoding end adopts mode illustrated in figures 1 and 2 to obtain gathers; If quantity is 2; Then only need bit of decoding, if quantity is 3, one or two bit of then need decoding.But under the error prone channels condition, possibly there is the situation of losing in reference frame.In case reference frame is lost, just might go wrong when then decoding end adopts mode shown in Figure 2 to search for time domain MVP, thereby cause decoding end candidate MVP number mistake.Suppose first order candidate MVP set that coding side obtains for adjacent candidate MVP} on time domain candidate MVP, the adjacent candidate MVP in a left side, spatial domain, the spatial domain, the MVP that therefrom selects is the adjacent candidate MVP in a left side, spatial domain, its index is 1, then corresponding index is encoded to 10.In decoding end, because reference frame is lost and can't be obtained time domain candidate MVP, the first order candidate MVP set that therefore obtains is { adjacent candidate MVP} on the adjacent candidate MVP in a left side, spatial domain, the spatial domain.Because the number of the candidate MVP that decoding end obtains is 2; Then decoding end only can be decoded to first bit " 1 " of index coding; And by error with the coding of second bit " 0 " as other part of current block; Thereby mistake occurs when causing decoding end that the encoding code stream except that the index coding is decoded, even cause whole decoding collapse, can't continue follow-up decode procedure.To the MV under the Merge pattern, also there is similar problem, repeat no more here.
Given this; Whether move candidate in the derivation ensemble of communication of the first order candidate that present embodiment obtains in can determining step 301 moves the number of derivation information greater than 2; If greater than 2; Then first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of this second level candidate moves the number of derivation information smaller or equal to 2.Therefore, adopt second level candidate to move the derivation ensemble of communication when current block is carried out encoding process, its motion derivation information index only needs a bit to represent.Table 2 is a kind of coded system of the candidate MVP of the technical scheme that adopts present embodiment.Need to prove; Be directed against the situation that only has a candidate MVP in the second level candidate MVP set in the table 2; The situation that is MVPNum=1 is also encoded, and it will be appreciated by persons skilled in the art that present embodiment is directed against this situation and also can encode.
Table 2
| MVPIdx | ?MVPNum＝1 | ?MVPNum＝2 |
| 0 | 0 | 0 |
| 1 | 1 |
Number with candidate MVP in the prior art is 3 to the maximum for instance; The number of candidate MVP in the first order candidate MVP set can only be { adjacent candidate MVP} on time domain candidate MVP, the adjacent candidate MVP in a left side, spatial domain, the spatial domain greater than 2 situation; Adopt the method for present embodiment, can these 3 candidate MVP be merged into two candidate MVP at the most.The mode of two candidate MVP of after the merging this shown in can employing table 2 carried out the index coding.Therefore, when decoding, decoding end can adopt these two the candidate MVP after above-mentioned identical processing mode is obtained merging.No matter whether reference frame is lost; Decoding end all only needs the index coding of a bit is decoded; And with remainder all as other part of the encoding code stream of current block; And can not occur thinking the part of index coding other part of the encoding code stream that is current block by mistake, thereby avoid wrong decoding.
Number with the candidate MV under the Merge pattern in the prior art is 5 to the maximum for instance; The number of candidate MV in the first order candidate MV set is more greater than 2 situation; For example can be { adjacent candidate MV, time domain candidate MV} on the adjacent candidate MV in a spatial domain left side, the spatial domain, can be { adjacent candidate MV} or the like under the adjacent candidate MV in a left side, spatial domain, time domain candidate MV, the upper right adjacent candidate MV in spatial domain, the left side, spatial domain; Adopt the method for present embodiment; Can these 3 or candidate MV more than 3 be merged into two candidate MV at the most, thereby only need a bit that the index of the MV that from second level candidate MV set, selects is encoded.
Need to prove; Present embodiment only is at most only to have 3 with candidate MVP in the prior art, and the candidate MV under the Merge pattern has 5 at most for example describes, and it will be understood by those skilled in the art that; Along with the development of technology, the number of candidate MVP or MV also can be more.In addition, present embodiment does not limit concrete merging mode, and those skilled in the art can adopt prior art or its combination that 3 or candidate more than 3 the derivation information of moving is merged into 1 or 2 candidates derivation information of moving.
In the present embodiment; Code device can move the number of derivation information the candidate during first order candidate moves the derivation ensemble of communication greater than under 2 the situation; First order candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving is processed at the most two candidates derivation information of moving through merging; To obtain the second level candidate derivation ensemble of communication of moving; And use this second level candidate derivation ensemble of communication of moving and carry out encoding process; Make this second level candidate is moved the motion derivation information index selected in the derivation ensemble of communication when encoding, need a bit at the most, thereby make decoding end move after the derivation ensemble of communication adopting above-mentioned identical merging processing mode to obtain second level candidate; No matter the situation that whether exists reference frame to lose; Decoding end only need be carried out decoding processing to the index coding of a bit, and can not occur thinking the part of index coding other part of the encoding code stream that is current block by mistake, thereby can avoid the encoding code stream of current block is decoded by error.
Adopt several kinds of concrete merging processing modes below, realize the technical scheme of code processing method embodiment shown in Figure 3.
Fig. 4 is the flow chart of video coding processing method embodiment two of the present invention, and as shown in Figure 4, the method for present embodiment can comprise:
With motion derivation information is that MVP is an example, and smaller or equal to 2, then code device does not carry out any union operation as if the total number of candidate MVP in the first order candidate MVP set of current block; If have time domain candidate MVP in the first order candidate MVP of the current block set, and the number of spatial domain candidate MVP is greater than 1, and then code device can merge whole spatial domain candidate MVP.For instance; Merging processing method in the present embodiment can be specially the spatial domain candidate MVP after the mean value of getting all spatial domain candidate MVP is handled as merging; Therefore; Can comprise two candidate MVP in this second level candidate MVP set, one is the spatial domain candidate MVP that merges after handling, and one is time domain candidate MVP.
In motion estimation process, code device can select a MVP to predict in second level candidate MVP set, and obtains final MV through motion search.Need to prove that if only have a unique candidate MVP in the second level candidate MVP set, then code device can directly utilize this unique candidate MVP to predict.
In the present embodiment; Code device can carry out encoding process to current block; Comprise the spatial domain candidate derivation information of moving in the derivation ensemble of communication if second level candidate moves, then code device can adopt a bit that the index of motion derivation information in this second level candidate moves the derivation ensemble of communication of selecting is encoded in the encoding code stream of this current block.Do not comprise any spatial domain candidate derivation information of moving in the derivation ensemble of communication if second level candidate moves, then code device is not encoded to this candidate derivation information index of moving.It is understandable that the benefit of not encoding is to practice thrift expense, simplify coded system.
It is understandable that to the MV under the Merge pattern, the difference of itself and MVP processing procedure is, motion estimation process is to choose the MV of the superior as current block among the MV that obtains of candidate MV and the motion search in the Merge pattern.
Present embodiment is on the basis of method embodiment shown in Figure 3; The merging processing mode of averaging through employing is merged into two at the most with first order candidate the move number of derivation information of candidate in the derivation information of moving, and makes code device can only adopt a bit that motion derivation information index is carried out encoding process at the most.
Fig. 5 is the flow chart of video coding processing method embodiment three of the present invention, and as shown in Figure 5, the method for present embodiment can comprise:
Be that MVP is an example with motion derivation information still, smaller or equal to 2, then code device does not carry out any union operation as if the total number of candidate MVP in the first order candidate MVP set of current block; If the number that does not have time domain candidate MVP and spatial domain candidate MVP in the first order candidate MVP of the current block set is greater than 2, then code device can be consolidated into two spatial domain candidate MVP at the most with spatial domain candidate MVP.For instance; Merging processing method in the present embodiment can be specially according to preset strategy; Move from said first order candidate and to select the derivation ensemble of communication at the most two spatial domain candidates to move derivation information as merging spatial domain candidate after the handling derivation information of moving; Wherein those skilled in the art can set up on their own this preset strategy, for example can select according to information such as the size of the value of each candidate MVP and/or positions.Therefore, can comprise one or two candidate spatial domains MVP in this second level candidate MVP set.Need to prove that among all embodiment of the present invention, preset strategy all can be for selecting according to information such as the size of the value of each candidate MVP and/or positions.
In motion estimation process, code device can select a MVP to predict in second level candidate MVP set, and obtains final MV through motion search.Need to prove that if only have a unique candidate MVP in the second level candidate MVP set, then code device can directly utilize this unique candidate MVP to predict.
Step 503, according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out encoding process.
In the present embodiment; Code device can carry out encoding process to current block; Comprise the spatial domain candidate derivation information of moving in the derivation ensemble of communication if second level candidate moves, then code device can adopt a bit that the index of motion derivation information in this second level candidate moves the derivation ensemble of communication of selecting is encoded in the encoding code stream of this current block.Do not comprise any spatial domain candidate derivation information of moving in the derivation ensemble of communication if second level candidate moves, then code device is not encoded to this candidate derivation information index of moving.
It is understandable that to the MV under the Merge pattern, the difference of itself and MVP processing procedure is, motion estimation process is to choose the MV of the superior as current block among the MV that obtains of candidate MV and the motion search in the Merge pattern.
Present embodiment is on the basis of method embodiment shown in Figure 3;, first order candidate do not have the time domain candidate derivation information of moving in moving the derivation ensemble of communication; According to preset strategy from the spatial domain candidate move the derivation information selected at the most two spatial domain candidates move derivation information as second level candidate candidate the derivation ensemble of communication derivation information of moving of moving, make code device can only adopt maximum bits that motion derivation information index is carried out encoding process.
Fig. 6 is the flow chart of video coding processing method embodiment four of the present invention, and as shown in Figure 6, the method for present embodiment can comprise:
Step 601, the first order candidate who the obtains current block derivation ensemble of communication of moving.
Step 602,, said first order candidate do not exist in the derivation ensemble of communication move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2 if moving; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving.
Before among several embodiment, at most only there is 1 in time domain motion derivation information, but present embodiment is to move to first order candidate only to exist in the derivation ensemble of communication move derivation information and time domain candidate of time domain candidate to move the number of derivation information greater than 2 situation.In this case; Code device can be consolidated at the most two time domain candidates derivation information of moving with the move number of derivation information of time domain candidate; Therefore, second level candidate moves and only comprises two the time domain candidates at the most that merge after handling derivation information of moving in the derivation ensemble of communication.
Step 603, according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out encoding process.
In the present embodiment, code device can carry out encoding process to current block, and code device can adopt a bit that the index of motion derivation information in this second level candidate moves the derivation ensemble of communication of selecting is encoded in the encoding code stream of this current block.
It both can be according to preset strategy that merging in the present embodiment is handled; Move from said first order candidate and to select the derivation ensemble of communication at the most two time domain candidates to move derivation information as merging time domain candidate after the handling derivation information of moving; Also can adopt other to merge processing mode, the mode of for example averaging.
Present embodiment is on the basis of method embodiment shown in Figure 3;, first order candidate only exist a plurality of time domain candidates to move under the situation of derivation information in moving the derivation ensemble of communication; Handle through merging; Make second level candidate the move number of derivation information of time domain candidate in the derivation ensemble of communication of moving have only two at the most, thereby make code device can only adopt a bit that motion derivation information index is carried out encoding process.
Fig. 7 is the flow chart of video coding processing method embodiment five of the present invention, and as shown in Figure 7, the method for present embodiment can comprise:
Present embodiment moves to first order candidate and has move derivation information and more than the move situation of derivation information of 1 time domain candidate of a spatial domain candidate in the derivation ensemble of communication.In this case; Code device can be consolidated into 1 time domain candidate derivation information of moving with the move number of derivation information of time domain candidate; Therefore, second level candidate moves and only comprises a time domain candidate who merges after handling derivation information and spatial domain candidate derivation information of moving of moving in the derivation ensemble of communication.
Merging in the present embodiment is handled; Both can be according to preset strategy; Moving from said first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information as merging time domain candidate after the handling derivation information of moving, can be again that a plurality of time domain candidates derivation information of moving is averaged to be consolidated into 1 time domain candidate derivation information of moving.
In the present embodiment, code device can carry out encoding process to current block, and code device can adopt a bit that the index of motion derivation information in this second level candidate moves the derivation ensemble of communication of selecting is encoded in the encoding code stream of this current block.
Present embodiment is on the basis of method embodiment shown in Figure 3;, first order candidate has move derivation information and move under the situation of derivation information of a spatial domain candidate in moving the derivation ensemble of communication more than 1 time domain candidate; Be consolidated into 1 time domain candidate derivation information of moving through derivation information that a plurality of time domain candidates are moved; Make second level candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving have only two, thereby make code device can only adopt a bit that motion derivation information index is carried out encoding process.
Fig. 8 is the flow chart of video coding processing method embodiment six of the present invention, and as shown in Figure 8, the method for present embodiment can comprise:
Step 801, the first order candidate who the obtains current block derivation ensemble of communication of moving.
Step 802, move derivation information and time domain candidate of spatial domain candidate in the derivation ensemble of communication moves the individual number average of derivation information greater than 1 if said first order candidate moves; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling.
Present embodiment moves in the derivation ensemble of communication to first order candidate and not only has the move derivation information but also have the move situation of derivation information more than 1 time domain candidate more than 1 spatial domain candidate.In this case, code device can be respectively handled spatial domain candidate derivation information and the time domain candidate derivation information of moving of moving.For the spatial domain candidate derivation information of moving; Code device can be consolidated into the spatial domain candidate derivation information of moving with a plurality of spatial domain candidates derivation information of moving; For the time domain candidate derivation information of moving; Code device can be consolidated into the time domain candidate derivation information of moving with a plurality of time domain candidates derivation information of moving; Therefore, the second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate after merge handling.
Merging in the present embodiment is handled; Both can be according to preset strategy; Move from first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information as merging time domain candidate after the handling derivation information of moving; Move from first order candidate and to select the derivation ensemble of communication spatial domain candidate to move derivation information as merging spatial domain candidate after the handling derivation information of moving; Can be again that a plurality of time domain candidates derivation information of moving is averaged being consolidated into 1 time domain candidate derivation information of moving, a plurality of spatial domain candidates derivation information of moving is averaged to be consolidated into 1 spatial domain candidate derivation information of moving.
Step 803, according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out encoding process.
In the present embodiment; Code device can carry out encoding process to current block; Comprise the spatial domain candidate derivation information of moving in the derivation ensemble of communication if second level candidate moves, then code device can adopt a bit that the index of motion derivation information in this second level candidate moves the derivation ensemble of communication of selecting is encoded in the encoding code stream of this current block.Do not comprise any spatial domain candidate derivation information of moving in the derivation ensemble of communication if second level candidate moves, then code device is not encoded to this candidate derivation information index of moving.
Present embodiment is on the basis of method embodiment shown in Figure 3;, first order candidate not only had move derivation information but also exist a plurality of time domain candidates to move under the situation of derivation information of a plurality of spatial domain candidates in moving the derivation ensemble of communication; Be consolidated into move derivation information and a plurality of spatial domain candidates derivation information of moving is consolidated into 1 spatial domain candidate derivation information of moving of 1 time domain candidate through derivation information that a plurality of time domain candidates are moved; Make second level candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving have only two, thereby make code device can only adopt maximum bits that motion derivation information index is carried out encoding process.
Fig. 9 is the flow chart of video coding processing method embodiment seven of the present invention, and as shown in Figure 9, the method for present embodiment can comprise:
Step 901, the first order candidate who the obtains current block derivation ensemble of communication of moving.
Step 902, comprise adjacent motion derivation information on spatial domain adjacent motion derivation information in a left side and/or the spatial domain in the derivation ensemble of communication at least, then from first order candidate moves the derivation ensemble of communication, only keep adjacent motion derivation information on spatial domain adjacent motion derivation information in a left side and/or the spatial domain as merging candidate after the handling derivation information of moving if said first order candidate moves.
Here; The adjacent motion derivation information in a spatial domain left side is taken from lateral coordinates left of current block (inferior like a positive left side, a left side) and in the movable information of encoding block, on the spatial domain adjacent motion derivation information take from lengthwise position above current block (as just go up, afterwards first-class) and in the movable information of encoding block.
Present embodiment moves to first order candidate and has the situation of multiple spatial domain motion derivation information and time domain motion derivation information combination in any in the derivation ensemble of communication.In this case, code device can be directly with adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving, and give up other candidate derivation information of moving.Therefore, the second level candidate derivation ensemble of communication of moving comprises following set: { having no motion derivation information }, { the adjacent motion derivation information in a left side, spatial domain }, adjacent motion derivation information, { adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain, the spatial domain } on the spatial domain.
Step 903, according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out encoding process.
In the present embodiment, code device can carry out encoding process to current block.Comprise at the most two candidates derivation information of moving in the derivation ensemble of communication because second level candidate moves, so code device can adopt a bit that the index of motion derivation information in this second level candidate moves the derivation ensemble of communication of selecting is encoded in the encoding code stream of this current block.For example adopt the coded system shown in the table 2.
Present embodiment is on the basis of method embodiment shown in Figure 3;, first order candidate exists in moving the derivation ensemble of communication under the situation of multiple spatial domain motion derivation information and time domain motion derivation information combination in any; Only keep adjacent motion derivation information on adjacent motion derivation information in a spatial domain left side and/or the spatial domain as second level candidate candidate in the derivation ensemble of communication derivation information of moving of moving, thereby make code device can only adopt a bit that motion derivation information index is carried out encoding process.
Figure 10 is the flow chart of video coding processing method embodiment eight of the present invention, and shown in figure 10, the method for present embodiment can comprise:
The particular content of the motion derivation information that present embodiment comprises in the derivation ensemble of communication no matter first order candidate moves; Candidate in the derivation ensemble of communication moves the number of derivation information greater than 1 as long as first order candidate moves, and just whole candidates derivation information of moving is consolidated into a motion derivation information.Therefore, the second level candidate derivation ensemble of communication of moving only comprises the motion derivation information after merge handling.
For instance; Can be for there be motion derivation information if first order candidate moves in the merging processing mode of present embodiment in the derivation ensemble of communication; The motion derivation information that then merges after handling is null vector; Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an odd number if first order candidate moves; Motion derivation information after then merge handling is said spatial domain candidate the move vectorial intermediate value of derivation information of derivation information and time domain candidate of moving; Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an even number if first order candidate moves, and the motion derivation information that then merges after handling is spatial domain candidate derivation information and the time domain candidate derivation information mean value of two the motion derivation information in centre after the sequence arrangement by size that moves that moves.
Because in the present embodiment; Second level candidate moves and has only the candidate derivation information of moving in the derivation ensemble of communication; Therefore, code device both can be encoded to this candidate derivation information index of moving, and again can be not this candidate derivation information index of moving not be encoded.It is understandable that the benefit of not encoding is to practice thrift expense, simplify coded system.
Present embodiment is on the basis of method embodiment shown in Figure 3; The motion derivation information that in first order candidate moves the derivation ensemble of communication, comprises is greater than under 1 the situation; Total movement derivation information is merged into a motion derivation information, and therefore, second level candidate moves and only comprises a motion derivation information in the derivation ensemble of communication; Thereby make code device both can not carry out the index coding, can only adopt a bit that motion derivation information index is carried out encoding process again.
Figure 11 is the flow chart of video coding processing method embodiment nine of the present invention, and shown in figure 11, the method for present embodiment can comprise:
With MVP for instance,, then be encoded to 0,, then be encoded to 1 if the MVP that selects is the spatial domain MVP in the second level candidate MVP set if the MVP that selects is the time domain MVP in the second level candidate MVP set.Therefore, present embodiment can only adopt a bit that the motion derivation information of selecting is carried out regular coding on the basis of previous embodiment, and the bit of coding has carried the information that MVP belongs to time domain or spatial domain.
Previous embodiment is only to adopting the MV under MVP or the Merge pattern that the process that current block carries out encoding process is specified.It is understandable that these two kinds of patterns also can merge use, before using, only need to set merging patterns flag MergeFlag and get final product.For instance; If current block has used the Merge pattern; And selected the MV of Merge pattern; Then code device can be set to 1 by MergeFlag, if the MV that current block does not use the Merge pattern or used the Merge pattern and selected motion search to obtain, then code device can be set to 0 by MergeFlag.MergeFlag=0, then code device need be encoded to the index of MVP in second candidate moves the derivation ensemble of communication of selecting for use, and its detailed process can adopt the process of above-mentioned arbitrary embodiment; If MergeFlag=1 then need encode to the index among the candidate MV of selected MV under the Merge pattern, its detailed process can also repeat no more with reference to the process of above-mentioned arbitrary embodiment here.
Figure 12 is the flow chart of video coding processing method embodiment ten of the present invention, and shown in figure 12, the method for present embodiment can comprise:
The step 121 of present embodiment is similar with the realization principle of the step 301 of method embodiment shown in Figure 3, repeats no more here.
There are time domain motion derivation information and spatial domain motion derivation information if the candidate moves in the derivation ensemble of communication, then from the candidate moves the derivation ensemble of communication, select a motion derivation information; The motion derivation information of selecting is carried out block code to be handled; The number of derivation information is irrelevant so that encoding code stream and said candidate move; Code device can carry out encoding process to the index of motion derivation information in said candidate moves the derivation ensemble of communication of selecting then; Said current block is carried out encoding process, obtain the encoding code stream of said current block, comprise the index coding in the said encoding code stream.To this situation, present embodiment can adopt the block code shown in the table 3.
Table 3
| MVPIdx | ?MVPNum＝1 | ?MVPNum＝2 | ?MVPNum＝3 |
| 0 | 0 | 0 | 0 |
| 1 | 10 | 10 | |
| 2 | 11 |
Can find out by table 3, this block code also can be so that encoding code stream and said candidate move the number of derivation information irrelevant.
Present embodiment moves to the candidate does not have move derivation information or only have the situation of a time domain motion derivation information of candidate in the derivation ensemble of communication; Code device does not carry out encoding process to motion derivation information index; Therefore, do not comprise motion derivation information index coding in the encoding code stream of current block.For the processing mode of index not being encoded; In decoding end; Because adopting the identical mode of coding side to obtain the candidate moves after the derivation ensemble of communication; Move derivation ensemble of communication or do not have candidate's derivation information of moving of this candidate, or have only the time domain candidate derivation information of moving is for there not being the move situation of derivation information of candidate; Decoding end can adopt any-mode to obtain the candidate derivation information of moving current block is carried out decoding processing; Present embodiment does not limit obtain manner, for having only the move situation of derivation information of a time domain candidate, decoding end can adopt this unique time domain candidate move derivation information or at this moment the territory candidate move under the prerequisite that derivation information can't obtain; Adopt alternate manner to obtain the candidate derivation information of moving, the encoding code stream of current block is carried out decoding processing get final product.The present embodiment code processing method has been practiced thrift the coding expense, and has simplified coded system.Need to prove that the motion derivation information in the present embodiment can also be the MV under MVP or the Merge pattern.
Figure 13 is the flow chart of video decode processing method embodiment one of the present invention, and shown in figure 13, the method for present embodiment can comprise:
The encoding code stream of step 131, reception current block.
For instance, decoding device can receive the encoding code stream of current block.This encoding code stream promptly can be to have carried out the encoding code stream that obtains after the technical scheme of coding embodiment shown in Figure 3.
The first order candidate that decoding device can the obtain current block derivation ensemble of communication of moving.
With motion derivation information be MVP for instance; Step 132 in the present embodiment can adopt existing techniques in realizing; Piece for example shown in Figure 1, that (F, G, H, I) selects first MV to use from top to bottom in all adjacent pieces of left side is with the MV of this piece adjacent MVP in a left side, spatial domain as current block; The piece that (A, B, C, D, E) selects first MV to use from right to left in all adjacent up pieces is with adjacent MVP on the spatial domain of MV as current block of this piece; As shown in Figure 2, in reference frame, select the reference block with the current block correspondence position, with the MV of this reference block center time domain MVP as current block.Owing to any MVP among the adjacent MVP on above-mentioned time domain MVP, the adjacent MVP in a left side, spatial domain, the spatial domain might not exist; Therefore, adopt the first order candidate MVP set of the method acquisition of prior art can comprise following any set form: { time domain MVP}, { the adjacent MVP} in a left side, spatial domain, { adjacent MVP}, { time domain MVP, the adjacent MVP} in a left side, spatial domain, { adjacent MVP}, { adjacent MVP} and { adjacent MVP} on time domain MVP, the left adjacent MVP in spatial domain, the spatial domain on the adjacent MVP in a left side, spatial domain, the spatial domain on time domain MVP, the spatial domain on the spatial domain.
With motion derivation information be under the Merge pattern MV for instance, the step 301 in the present embodiment also can adopt existing techniques in realizing.Candidate MV possibly comprise on the adjacent MV in a spatial domain left side, the spatial domain adjacent MV under adjacent MV, time domain MV, the upper right adjacent MV in spatial domain, the left side, spatial domain; Therefore; With above-mentioned candidate MVP similarly, first order candidate MV set also can include some kinds of set forms of at least one composition among above-mentioned five kinds of candidate MV.
It will be appreciated by persons skilled in the art that step 132 also can not adopt existing techniques in realizing, for example, for candidate MVP, other time domain candidate MVP more than other spatial domain candidate MVP perhaps comprised more more than it also possibly comprise more; For the candidate MV under the Merge pattern, other time domain candidate MV more than other spatial domain candidate MV perhaps comprised more more than it also possibly comprise more.The first order candidate that present embodiment does not obtain decoding device the move particular content of derivation information of the candidate that comprises in the derivation ensemble of communication that moves limits, and gets final product so long as can carry out the motion derivation information of interframe prediction decoding.
At coding side; If only comprise 1 candidate MVP in the first order candidate MVP set of adopting mode illustrated in figures 1 and 2 to obtain; It is the situation of corresponding MVPNum=1; Then encoding device can not encoded to the index of this candidate MVP, and this unique candidate MVP possibly be any among the adjacent candidate MVP on time domain candidate MVP, the spatial domain adjacent candidate MVP in a left side and the spatial domain; If comprise 2 candidate MVP in the first order candidate MVP set of adopting mode illustrated in figures 1 and 2 to obtain; It is the situation of corresponding MVPNum=2; The corresponding MVPIdx of the 1st candidate MVP in the first order candidate MVP set is 0, and its corresponding index is encoded to 0, and the corresponding MVPIdx of the 2nd candidate MVP in the first order candidate MVP set is 1; Its corresponding index is encoded to 1, and these two candidate MVP possibly be any two among the adjacent candidate MVP on time domain candidate MVP, the spatial domain adjacent candidate MVP in a left side and the spatial domain; If comprise 3 candidate MVP in the first order candidate MVP set of adopting mode illustrated in figures 1 and 2 to obtain; It is the situation of corresponding MVPNum=3; The corresponding MVPIdx of the 1st candidate MVP in the first order candidate MVP set is 0, and its corresponding index is encoded to 0, and the corresponding MVPIdx of the 2nd candidate MVP in the first order candidate MVP set is 1; Its corresponding index is encoded to 10; The corresponding MVPIdx of the 3rd candidate MVP in the first order candidate MVP set is 2, and its corresponding index is encoded to 11, and these 3 candidate MVP are adjacent candidate MVP on time domain candidate MVP, the spatial domain adjacent candidate MVP in a left side and the spatial domain.Hence one can see that, and when the coding method of employing prior art was encoded to the index of these 3 candidate MVP in first order candidate MVP set, this index coding possibly be a bit; Also possibly be two bits, therefore, decoding end is when adopting prior art that the index coding is carried out decoding processing; Both possibly need bit of decoding, two bits of also need decoding, the quantity of candidate MVP during this first order candidate MVP that depends on that decoding end adopts mode illustrated in figures 1 and 2 to obtain gathers; If quantity is 2; Then only need bit of decoding, if quantity is 3, one or two bit of then need decoding.But under the error prone channels condition, possibly there is the situation of losing in reference frame.In case reference frame is lost, just might go wrong when then decoding end adopts mode shown in Figure 2 to search for time domain MVP, thereby cause decoding end candidate MVP number mistake.Suppose first order candidate MVP set that coding side obtains for adjacent candidate MVP} on time domain candidate MVP, the adjacent candidate MVP in a left side, spatial domain, the spatial domain, the MVP that therefrom selects is the adjacent candidate MVP in a left side, spatial domain, its index is 1, then corresponding index is encoded to 10.In decoding end, because reference frame is lost and can't be obtained time domain candidate MVP, the first order candidate MVP set that therefore obtains is { adjacent candidate MVP} on the adjacent candidate MVP in a left side, spatial domain, the spatial domain.Because the number of the candidate MVP that decoding end obtains is 2; Then decoding end only can be decoded to first bit " 1 " of index coding; And by error with the coding of second bit " 0 " as other part of current block; Thereby mistake occurs when causing decoding end that the encoding code stream except that the index coding is decoded, even cause whole decoding collapse, can't continue follow-up decode procedure.To the MV under the Merge pattern, also there is similar problem, repeat no more here.
Given this; With the technical scheme of encoding process embodiment shown in Figure 3 correspondingly; In the present embodiment; Whether move candidate in the derivation ensemble of communication of the first order candidate that decoding device obtains in can determining step 132 moves the number of derivation information greater than 2, if greater than 2, then first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of this second level candidate moves the number of derivation information smaller or equal to 2.Therefore, adopt second level candidate to move the derivation ensemble of communication when current block is carried out decoding processing, its motion derivation information index only accounts for a bit, and decoding device can be only carry out decoding processing to the index coding of a bit.
Number with candidate MVP in the prior art is 3 to the maximum for instance; The number of candidate MVP in the first order candidate MVP set can only be { adjacent candidate MVP} on time domain candidate MVP, the adjacent candidate MVP in a left side, spatial domain, the spatial domain greater than 2 situation; Adopt the method for present embodiment, decoding device can be merged into two candidate MVP at the most with these 3 candidate MVP.Because coding side is also adopted in a like fashion the candidate MVP in the first order candidate MVP set merged and is processed at the most two candidate MVP and only to need to adopt a bit index of the MVP that selects is encoded; Therefore; Correspondingly; In decoding end, decoding device can adopt table 2 these two candidate MVP after merging and select MVP, and utilizes this MVP that the encoding code stream of current block is carried out decoding processing.Therefore; No matter whether reference frame is lost; Decoding end all only needs the index coding of a bit is decoded; And with remainder all as other part of the encoding code stream of current block, and can not occur thinking the part of index coding other part of the encoding code stream that is current block by mistake, thereby avoid wrong decoding.
Number with the candidate MV under the Merge pattern in the prior art is 5 to the maximum for instance; The number of candidate MV in the first order candidate MV set is more greater than 2 situation; For example can be { adjacent candidate MV, time domain candidate MV} on the adjacent candidate MV in a spatial domain left side, the spatial domain, can be { adjacent candidate MV} or the like under the adjacent candidate MV in a left side, spatial domain, time domain candidate MV, the upper right adjacent candidate MV in spatial domain, the left side, spatial domain; Adopt the coding/decoding method of present embodiment; Can these 3 or candidate MV more than 3 be merged into two candidate MV at the most; Thereby only need be through bit of decoding, the MV that can from second level candidate MV set, select carries out decoding processing with the encoding code stream to current block.
Need to prove; Present embodiment only is at most only to have 3 with candidate MVP in the prior art, and the candidate MV under the Merge pattern has 5 at most for example describes, and it will be understood by those skilled in the art that; Along with the development of technology, the number of candidate MVP or MV also can be more.In addition, present embodiment does not limit concrete merging mode, and those skilled in the art can adopt prior art or its combination that 3 or candidate more than 3 the derivation information of moving is merged into 1 or 2 candidates derivation information of moving.
In addition; Need to prove; The foregoing description is that example describes with table 2 only, it will be appreciated by persons skilled in the art that for having only the move situation of derivation information of a candidate after merge handling at the most; The code device of coding side can be not carry out encoding process to this candidate derivation information of moving; Correspondingly, have only the candidate derivation information of moving at the most when decoding end adopts identical merging processing mode, then the decoding device of decoding end can not carry out decoding processing to the index coding yet yet; And directly adopt this candidate derivation information of moving the encoding code stream of current block to be carried out decoding processing or adopt alternate manner to obtain the candidate derivation information of moving, the encoding code stream of current block is carried out decoding processing get final product
In the present embodiment; The decoding device of decoding end can move the number of derivation information the candidate during first order candidate moves the derivation ensemble of communication greater than under 2 the situation; First order candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving is processed at the most two candidates derivation information of moving through merging; Obtaining the second level candidate derivation ensemble of communication of moving, and use this second level candidate derivation ensemble of communication of moving the encoding code stream of current block is carried out decoding processing, when decoding; The decoding device bit of only need decoding; No matter the situation that whether exists reference frame to lose, decoding end all can not occur thinking the part of index coding other part of the encoding code stream that is current block by mistake, thereby can avoid the encoding code stream of current block is decoded by error.
Figure 14 is the flow chart of video decode processing method embodiment two of the present invention, and shown in figure 14, the method for present embodiment can comprise:
The encoding code stream of step 141, reception current block.
Step 142, the first order candidate who the obtains current block derivation ensemble of communication of moving.
The realization principle of step 142 and the realization principle of above-mentioned steps 132 are similar, repeat no more here.
Step 143,, said first order candidate exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1 if moving; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving.
With motion derivation information is that MVP is an example, and smaller or equal to 2, then decoding device does not carry out any union operation as if the total number of candidate MVP in the first order candidate MVP set of current block; If have time domain candidate MVP in the first order candidate MVP of the current block set, and the number of spatial domain candidate MVP is greater than 1, and then can to merge whole spatial domain candidate MVP be a spatial domain candidate MVP to decoding device.For instance; Merging processing method in the present embodiment can be specially the spatial domain candidate MVP after the mean value of getting all spatial domain candidate MVP is handled as merging; Therefore, can comprise two candidate MVP in this second level candidate MVP set, wherein; One is the spatial domain candidate MVP that merges after handling, and one is time domain candidate MVP.
Step 144, according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out decoding processing.
In the present embodiment; Decoding device can carry out decoding processing to current block; If moving, said second level candidate comprises the spatial domain candidate derivation information of moving in the derivation ensemble of communication; Then the motion derivation information index coding to said current block carries out decoding processing, obtains index information, from said second level candidate moves the derivation ensemble of communication, selects motion derivation information according to said index information; If moving, said second level candidate do not comprise any spatial domain candidate derivation information of moving in the derivation information, then according to only there being candidate's derivation information of moving that the encoding code stream of said current block is carried out decoding processing.
It is understandable that to the MV under the Merge pattern, itself and MVP processing procedure are similar, repeat no more here.
Present embodiment is on the basis of method embodiment shown in Figure 13; The merging processing mode of averaging through employing is merged into two at the most with first order candidate the move number of derivation information of candidate in the derivation information of moving, and makes decoding device can only adopt maximum bits that motion derivation information index is carried out decoding processing.
Figure 15 is the flow chart of video decode processing method embodiment three of the present invention, and shown in figure 15, the method for present embodiment can comprise:
The encoding code stream of step 151, reception current block.
Be that MVP is an example with motion derivation information still, smaller or equal to 2, then decoding device does not carry out any union operation as if the total number of candidate MVP in the first order candidate MVP set of current block; If the number that does not have time domain candidate MVP and spatial domain candidate MVP in the first order candidate MVP of the current block set is greater than 2, then decoding device can be consolidated into two spatial domain candidate MVP at the most with spatial domain candidate MVP.For instance; Merging processing method in the present embodiment can be specially according to preset strategy; Move from said first order candidate and to select the derivation ensemble of communication at the most two spatial domain candidates to move derivation information as merging spatial domain candidate after the handling derivation information of moving; Wherein those skilled in the art can set up on their own this preset strategy, for example can select according to information such as the size of the value of each candidate MVP and/or positions.Therefore, can comprise one or two candidate spatial domains MVP in this second level candidate MVP set.
In the present embodiment; Decoding device can carry out decoding processing to current block; If moving, said second level candidate comprises the spatial domain candidate derivation information of moving in the derivation ensemble of communication; Then the motion derivation information index coding to said current block carries out decoding processing, obtains index information, from said second level candidate moves the derivation ensemble of communication, selects motion derivation information according to said index information; Do not comprise any spatial domain candidate derivation information of moving in the derivation information if said second level candidate moves, then the encoding code stream of said current block is carried out decoding processing according to only candidate derivation information of moving.
It is understandable that to the MV under the Merge pattern, itself and MVP processing procedure are similar, repeat no more here.
Present embodiment is on the basis of method embodiment shown in Figure 13;, first order candidate do not have the time domain candidate derivation information of moving in moving the derivation ensemble of communication; According to preset strategy from the spatial domain candidate move the derivation information selected at the most two spatial domain candidates move derivation information as second level candidate candidate the derivation ensemble of communication derivation information of moving of moving, make decoding device can only adopt maximum bits that motion derivation information index is carried out decoding processing.
Figure 16 is the flow chart of video decode processing method embodiment four of the present invention, and shown in figure 16, the method for present embodiment can comprise:
The encoding code stream of step 161, reception current block.
Before among several embodiment, at most only there is 1 in time domain motion derivation information, but present embodiment is to move to first order candidate only to exist in the derivation ensemble of communication move derivation information and time domain candidate of time domain candidate to move the number of derivation information greater than 2 situation.In this case; Decoding device can be consolidated at the most two time domain candidates derivation information of moving with the move number of derivation information of time domain candidate; Therefore, second level candidate moves and only comprises two the time domain candidates at the most that merge after handling derivation information of moving in the derivation ensemble of communication.
In the present embodiment; Decoding device can carry out decoding processing to current block; Motion derivation information index coding to said current block carries out decoding processing; Obtain index information, move from said second level candidate according to said index information and select motion derivation information the derivation ensemble of communication, and the encoding code stream of said current block is carried out decoding processing according to the motion derivation information of selecting.
It both can be according to preset strategy that merging in the present embodiment is handled; Move from said first order candidate and to select the derivation ensemble of communication at the most two time domain candidates to move derivation information as merging time domain candidate after the handling derivation information of moving; Also can adopt other to merge processing mode, the mode of for example averaging.
Present embodiment is on the basis of method embodiment shown in Figure 13;, first order candidate only exist a plurality of time domain candidates to move under the situation of derivation information in moving the derivation ensemble of communication; Handle through merging; Make second level candidate the move number of derivation information of time domain candidate in the derivation ensemble of communication of moving have only two at the most, thereby make the decoding device bit derivation information index that obtains moving of can only decoding.
Figure 17 is the flow chart of video decode processing method embodiment five of the present invention, and shown in figure 17, the method for present embodiment can comprise:
The encoding code stream of step 171, reception current block.
Present embodiment moves to first order candidate and has move derivation information and more than the move situation of derivation information of 1 time domain candidate of a spatial domain candidate in the derivation ensemble of communication.In this case; Decoding device can be consolidated into 1 time domain candidate derivation information of moving with the move number of derivation information of time domain candidate; Therefore, second level candidate moves and only comprises a time domain candidate who merges after handling derivation information and spatial domain candidate derivation information of moving of moving in the derivation ensemble of communication.
Merging in the present embodiment is handled; Both can be according to preset strategy; Moving from said first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information as merging time domain candidate after the handling derivation information of moving, can be again that a plurality of time domain candidates derivation information of moving is averaged to be consolidated into 1 time domain candidate derivation information of moving.
In the present embodiment; Decoding device can carry out decoding processing to current block; Motion derivation information index coding to said current block carries out decoding processing; Obtain index information, move from said second level candidate according to said index information and select motion derivation information the derivation ensemble of communication, and the encoding code stream of said current block is carried out decoding processing according to the motion derivation information of selecting.
Present embodiment is on the basis of method embodiment shown in Figure 13;, first order candidate has move derivation information and move under the situation of derivation information of a spatial domain candidate in moving the derivation ensemble of communication more than 1 time domain candidate; Be consolidated into 1 time domain candidate derivation information of moving through derivation information that a plurality of time domain candidates are moved; Make second level candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving have only two, thereby make the decoding device bit derivation information index that obtains moving of can only decoding.
Figure 18 is the flow chart of video decode processing method embodiment six of the present invention, and shown in figure 18, the method for present embodiment can comprise:
The encoding code stream of step 181, reception current block.
Present embodiment moves in the derivation ensemble of communication to first order candidate and not only has the move derivation information but also have the move situation of derivation information more than 1 time domain candidate more than 1 spatial domain candidate.In this case, decoding device can be respectively handled spatial domain candidate derivation information and the time domain candidate derivation information of moving of moving.For the spatial domain candidate derivation information of moving; Decoding device can be consolidated into the spatial domain candidate derivation information of moving with a plurality of spatial domain candidates derivation information of moving; For the time domain candidate derivation information of moving; Decoding device can be consolidated into the time domain candidate derivation information of moving with a plurality of time domain candidates derivation information of moving; Therefore, the second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate after merge handling.
Merging in the present embodiment is handled; Both can be according to preset strategy; Move from first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information as merging time domain candidate after the handling derivation information of moving; Move from first order candidate and to select the derivation ensemble of communication spatial domain candidate to move derivation information as merging spatial domain candidate after the handling derivation information of moving; Can be again that a plurality of time domain candidates derivation information of moving is averaged being consolidated into 1 time domain candidate derivation information of moving, a plurality of spatial domain candidates derivation information of moving is averaged to be consolidated into 1 spatial domain candidate derivation information of moving.
In the present embodiment; Decoding device can carry out decoding processing to current block; Comprise the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said second level candidate moves, then the motion derivation information index coding to said current block carries out decoding processing, obtains index information; Move from said second level candidate according to said index information and to select motion derivation information the derivation ensemble of communication, and the encoding code stream of said current block is carried out decoding processing according to the motion derivation information of selecting; Do not comprise any candidate derivation information of moving in the derivation information if said second level candidate moves, then the encoding code stream of said current block is carried out decoding processing according to only candidate derivation information of moving.
Present embodiment is on the basis of method embodiment shown in Figure 13;, first order candidate not only had move derivation information but also exist a plurality of time domain candidates to move under the situation of derivation information of a plurality of spatial domain candidates in moving the derivation ensemble of communication; Be consolidated into move derivation information and a plurality of spatial domain candidates derivation information of moving is consolidated into 1 spatial domain candidate derivation information of moving of 1 time domain candidate through derivation information that a plurality of time domain candidates are moved; Make second level candidate the move number of derivation information of candidate in the derivation ensemble of communication of moving have only two, thereby make decoding device can only adopt maximum bits that motion derivation information index is carried out decoding processing.
Figure 19 is the flow chart of video decode processing method embodiment seven of the present invention, and shown in figure 19, the method for present embodiment can comprise:
The encoding code stream of step 191, reception current block.
Here, the adjacent motion derivation information in a spatial domain left side is taken from the position at the left of current block and the movable information of decoding block, and adjacent motion derivation information is taken from above current block and the movable information of decoding block of position on the spatial domain.
Present embodiment moves to first order candidate and has the situation of multiple spatial domain motion derivation information and time domain motion derivation information combination in any in the derivation ensemble of communication.In this case, decoding device can be directly with adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving, and give up other candidate derivation information of moving.Therefore, the second level candidate derivation ensemble of communication of moving comprises following set: { having no motion derivation information }, { the adjacent motion derivation information in a left side, spatial domain }, adjacent motion derivation information, { adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain, the spatial domain } on the spatial domain.
In the present embodiment, decoding device can carry out decoding processing to current block.Because moving, second level candidate comprises at the most two candidates derivation information of moving in the derivation ensemble of communication, so the decoding device bit derivation information index that obtains moving of can only decoding.
Present embodiment is on the basis of method embodiment shown in Figure 13;, first order candidate exists in moving the derivation ensemble of communication under the situation of multiple spatial domain motion derivation information and time domain motion derivation information combination in any; Only keep adjacent motion derivation information on adjacent motion derivation information in a spatial domain left side and/or the spatial domain as second level candidate candidate in the derivation ensemble of communication derivation information of moving of moving, thereby make the decoding device bit derivation information index that obtains moving of can only decoding.
Figure 20 is the flow chart of video decode processing method embodiment eight of the present invention, and shown in figure 20, the method for present embodiment can comprise:
The encoding code stream of step 201, reception current block.
The particular content of the motion derivation information that present embodiment comprises in the derivation ensemble of communication no matter first order candidate moves; Candidate in the derivation ensemble of communication moves the number of derivation information greater than 1 as long as first order candidate moves, and just whole candidates derivation information of moving is consolidated into a motion derivation information.Therefore, the second level candidate derivation ensemble of communication of moving only comprises the motion derivation information after merge handling.
For instance; Can be for there be motion derivation information if first order candidate moves in the merging processing mode of present embodiment in the derivation ensemble of communication; The motion derivation information that then merges after handling is null vector; Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an odd number if first order candidate moves; Motion derivation information after then merge handling is said spatial domain candidate the move vectorial intermediate value of derivation information of derivation information and time domain candidate of moving; Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an even number if first order candidate moves, and the motion derivation information that then merges after handling is spatial domain candidate derivation information and the time domain candidate derivation information mean value of two the motion derivation information in centre after the sequence arrangement by size that moves that moves.
Because in the present embodiment, second level candidate moves and has only the candidate derivation information of moving in the derivation ensemble of communication, and therefore, decoding device both can be decoded to the index coding, can the index coding not decoded again.
Present embodiment is on the basis of method embodiment shown in Figure 13; The motion derivation information that in first order candidate moves the derivation ensemble of communication, comprises is greater than under 1 the situation; Total movement derivation information is merged into a motion derivation information, and therefore, second level candidate moves and only comprises a motion derivation information in the derivation ensemble of communication; Thereby make decoding device both can not carry out the decoding processing of index coding, can only encode again and carry out decoding processing the index of a bit.
Figure 21 is the flow chart of video decode processing method embodiment nine of the present invention, and shown in figure 21, the method for present embodiment can comprise:
The encoding code stream of step 211, reception current block.
In the present embodiment, decoding device can carry out decoding processing to current block.Because moving, second level candidate comprises at the most two candidates derivation information of moving in the derivation ensemble of communication, so the decoding device bit derivation information index that obtains moving of can only decoding.
With MVP for instance,, can know that then coding side selected time domain MVP, and the mark current block uses is time domain MVP if decoding MVP index is 0; If decoding MVP index is 1, can know that then coding side selected spatial domain MVP, and the mark current block uses is spatial domain MVP.
Previous embodiment is only to adopting the MV under MVP or the Merge pattern that the process that current block carries out decoding processing is specified.It is understandable that these two kinds of patterns also can merge use, before using, only need to judge that merging patterns flag MergeFlag gets final product.For instance, if MergeFlag=0, then decoding device need be decoded to the index coding of MVP in second candidate moves the derivation ensemble of communication of selecting for use, and its detailed process can adopt the process of above-mentioned arbitrary embodiment; If MergeFlag=1, then decoding device need be decoded to the coding of the index among the candidate MV of selected MV under the Merge pattern, and its detailed process can also repeat no more with reference to the process of above-mentioned arbitrary embodiment here.
Figure 22 is the flow chart of video decode processing method embodiment ten of the present invention, and shown in figure 22, the method for present embodiment can comprise:
The encoding code stream of step 221, reception current block.
The step 222 of present embodiment is similar with the realization principle of the step 131 of method embodiment shown in Figure 13, repeats no more here.
Have time domain motion derivation information and spatial domain motion derivation information if said candidate moves in the derivation information, then decoding device can carry out decoding processing to first of motion derivation information index coding, obtains first information; If corresponding first index of said first information, then confirm the said candidate motion derivation information in the derivation information of moving according to said first index; If corresponding second index of said first information; Then the next bit to said index coding carries out decoding processing; Obtain next bit information; If said first information and next bit information correspondence second index, then confirm the said candidate motion derivation information in the derivation information of moving according to said second index; And the like, be worth shared bit number up to motion derivation information index and reach maximum.
To this situation, the block code mode of present embodiment shown in can employing table 3 carried out decoding processing.For instance, if index is encoded to 0, then the index of MV is 0, if index is encoded to 1; Then continue 1 bit of decoding, if next bit is 0, then the index of MV is 1; If next bit is 1, then the index of MV is 2, by that analogy up to resolving to 0 or the MV index maximum number bits that possibly take.Do not having candidate MV or only existing under the situation of time domain candidate MV, decoding device can not decoded, and the index of MV is 0.Decoding device can be selected corresponding MV according to the index that parses in candidate MV.
Present embodiment moves to the candidate does not have move derivation information or only have the situation of a time domain motion derivation information of candidate in the derivation ensemble of communication; Decoding device does not carry out decoding processing to motion derivation information index coding; For there not being the move situation of derivation information of candidate; Decoding device can adopt any-mode to obtain the candidate derivation information of moving current block is carried out decoding processing; Present embodiment does not limit obtain manner, for having only the move situation of derivation information of a time domain candidate, decoding device can adopt this unique time domain candidate move derivation information or at this moment the territory candidate move under the prerequisite that derivation information can't obtain; Adopt alternate manner to obtain the candidate derivation information of moving, the encoding code stream of current block is carried out decoding processing get final product.Need to prove that the motion derivation information in the present embodiment can also be the MV under MVP or the Merge pattern.
Figure 23 is the structural representation of code device embodiment one of the present invention; Shown in figure 23; The device of present embodiment can comprise: first acquisition module 11, first merges the processing module 12 and the first encoding process module 13; Wherein, first acquisition module 11 first order candidate that is used to the to obtain current block derivation ensemble of communication of moving; The candidate of the first derivation ensemble of communication if said first order candidate moves that merges that processing module 12 is used for moves the number of derivation information greater than 2; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2; The first encoding process module 13 is used for candidate that the derivation ensemble of communication the comprises derivation information of moving of moving according to said second level candidate, and said current block is carried out encoding process.
The device of present embodiment can be used to carry out the method for method embodiment shown in Figure 3, and its realization principle and technique effect are similar, repeat no more here.
On the basis of apparatus structure shown in Figure 23, in another embodiment of code device, first merges processing module 12 can specifically be used for:
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving; Perhaps,
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 2; Then the said spatial domain candidate derivation information of moving is consolidated at the most two spatial domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the spatial domain candidates at the most that merge after handling derivation information of moving; Perhaps,
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving; Perhaps,
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and time domain candidate of a spatial domain candidate to move the number of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises a time domain candidate who merges after handling derivation information and the said spatial domain candidate derivation information of moving of moving; Perhaps,
Move derivation information and time domain candidate of spatial domain candidate in the derivation ensemble of communication moves the individual number average of derivation information greater than 1 if said first order candidate moves; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling; Perhaps,
Move from first order candidate and to select the derivation ensemble of communication adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving; Perhaps,
Said first order candidate whole candidates in derivation information derivation information of moving of moving is consolidated into a motion derivation information.
In addition; The first encoding process module 13 can be used for specifically that the derivation ensemble of communication comprises two candidates derivation information of moving if said second level candidate moves; Then, said second level candidate selects a motion derivation information from moving the derivation ensemble of communication; Said current block is carried out encoding process, obtain the encoding code stream of said current block, comprise the index coding of motion derivation information in said second level candidate moves the derivation ensemble of communication of selection in the said encoding code stream; Comprise the candidate derivation information of moving in the derivation ensemble of communication if said second level candidate moves, then said current block is carried out encoding process, do not comprise the index coding in the encoding code stream of said current block.
The processing procedure of the merging of first in present embodiment processing module 12 and the processing procedure of the first encoding process module 13 all can be specifically referring to the implementation procedures of Fig. 4～11 illustrated embodiments, and its realization principle and technique effect are similar, repeat no more here.
Figure 24 is the structural representation of code device embodiment two of the present invention; Shown in figure 24; The device of present embodiment can comprise: second acquisition module 21 and the second encoding process module 22, wherein, the candidate that second acquisition module 21 is used to the to obtain current block derivation ensemble of communication of moving; The second encoding process module 22 is used for that there is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves; Then said current block is carried out encoding process; Obtain the encoding code stream of said current block, do not comprise motion derivation information index coding in the said encoding code stream.
The device of present embodiment can be used to carry out the method for method embodiment shown in Figure 12, and its realization principle and technique effect are similar, repeat no more here.
Figure 25 is the structural representation of decoding device embodiment one of the present invention; Shown in figure 25; The device of present embodiment can comprise: first receiver module 31, the 3rd acquisition module 32, second merge the processing module 33 and first decoding process module 34; Wherein, first receiver module 31 is used to receive the encoding code stream of current block; The first order candidate that the 3rd acquisition module 32 is used to the to obtain said current block derivation ensemble of communication of moving; The candidate of the second derivation ensemble of communication if said first order candidate moves that merges that processing module 33 is used for moves the number of derivation information greater than 2; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2; First decoding process module 34 carries out decoding processing according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving to the encoding code stream of said current block.
The device of present embodiment can be used to carry out the method for method embodiment shown in Figure 13, and its realization principle and technique effect are similar, repeat no more here.
On the basis of apparatus structure shown in Figure 25, in another embodiment of decoding device, second merges processing module 33 can specifically be used for:
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving; Perhaps,
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 2; Then the said spatial domain candidate derivation information of moving is consolidated at the most two spatial domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the spatial domain candidates at the most that merge after handling derivation information of moving; Perhaps,
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving; Perhaps,
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and time domain candidate of a spatial domain candidate to move the number of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises a time domain candidate who merges after handling derivation information and the said spatial domain candidate derivation information of moving of moving; Perhaps,
Move derivation information and time domain candidate of spatial domain candidate in the derivation ensemble of communication moves the individual number average of derivation information greater than 1 if said first order candidate moves; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling; Perhaps,
Move from first order candidate and to select the derivation ensemble of communication adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving; Perhaps,
Said first order candidate whole candidates in derivation information derivation information of moving of moving is consolidated into a motion derivation information.
In addition; First decoding process module 34 is used for specifically that the derivation ensemble of communication comprises two candidates derivation information of moving if said second level candidate moves; Then the motion derivation information index coding to said current block carries out decoding processing; Obtain index information, move from said second level candidate according to said index information and select motion derivation information the derivation ensemble of communication, and the encoding code stream of said current block is carried out decoding processing according to the motion derivation information of selecting; Comprise the candidate derivation information of moving in the derivation information if said second level candidate moves, then the encoding code stream of said current block is carried out decoding processing according to this candidate derivation information of moving.
The processing procedure of the merging of second in present embodiment processing module 33 and the processing procedure of first decoding process module 34 all can be specifically referring to the implementation procedures of Figure 14～21 illustrated embodiments, and its realization principle and technique effect are similar, repeat no more here.
Figure 26 is the structural representation of decoding device embodiment two of the present invention; Shown in figure 26; The device of present embodiment can comprise: second receiver module 41, the 4th acquisition module 42 and second decoding process module 43, and wherein, second receiver module 41 is used to receive the encoding code stream of current block; The candidate that the 4th acquisition module 42 is used to the to obtain current block derivation ensemble of communication of moving; Second decoding process module 43 is used for that there is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, and then according to the non-spatial domain candidate derivation information of moving the encoding code stream of said current block is carried out decoding processing.
The device of present embodiment can be used to carry out the method for method embodiment shown in Figure 22, and its realization principle and technique effect are similar, repeat no more here.
The technology that the embodiment of the invention provides can be applied in the digital processing field, and through encoder, decoder is realized.Video encoder, decoder are widely used in various communication apparatus or the electronic equipment, for example: DTV, STB, WMG, mobile phone; Wireless device, personal digital assistant (PDA), hand-hold type or portable computer, GPS receiver/omniselector; Camera, video player, video camera; Video tape recorder, watch-dog, video conference and videophone device or the like.Comprise processor in this kind equipment, memory, and the interface of transmission data.Video Codec can be directly by digital circuit or chip for example DSP (digital signal processor) realize, perhaps drive the flow process in the processor software code and realize by software code.
One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be accomplished through the relevant hardware of program command; Aforesaid program can be stored in the computer read/write memory medium; This program the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CD.
What should explain at last is: above embodiment is only in order to explaining technical scheme of the present invention, but not to its restriction; Although with reference to previous embodiment the present invention has been carried out detailed explanation, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment put down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these are revised or replacement, do not make the spirit and the scope of the essence disengaging various embodiments of the present invention technical scheme of relevant art scheme.
Claims (46)
1. a video coding processing method is characterized in that, comprising:
The first order candidate who the obtains current block derivation ensemble of communication of moving;
Candidate in the derivation ensemble of communication moves the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
According to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, said current block is carried out encoding process.
2. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving.
3. method according to claim 2 is characterized in that, said merging is handled, and comprising:
Move whole spatial domain candidates in the derivation ensemble of communication of said first order candidate are moved the mean value of derivation information as merging spatial domain candidate after the handling derivation information of moving.
4. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 2; Then the said spatial domain candidate derivation information of moving is consolidated at the most two spatial domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the spatial domain candidates at the most that merge after handling derivation information of moving.
5. method according to claim 4 is characterized in that, said merging is handled and comprised:
According to preset strategy, move from said first order candidate and to select the derivation ensemble of communication at the most two spatial domain candidates to move derivation information as merging spatial domain candidate after the handling derivation information of moving.
6. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving.
7. method according to claim 6 is characterized in that, said merging is handled, and comprising:
According to preset strategy, move from said first order candidate and to select the derivation ensemble of communication at the most two time domain candidates to move derivation information as merging time domain candidate after the handling derivation information of moving.
8. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and time domain candidate of a spatial domain candidate to move the number of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises a time domain candidate who merges after handling derivation information and the said spatial domain candidate derivation information of moving of moving.
9. method according to claim 8 is characterized in that, said merging is handled, and comprising:
According to preset strategy, move from said first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information as merging time domain candidate after the handling derivation information of moving.
10. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
Move derivation information and time domain candidate of spatial domain candidate in the derivation ensemble of communication moves the individual number average of derivation information greater than 1 if said first order candidate moves; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling.
11. method according to claim 10 is characterized in that, said merging is handled, and comprising:
According to preset strategy; Move from said first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information, from said first order candidate moves the derivation ensemble of communication, select a spatial domain candidate to move derivation information as merging spatial domain candidate after the handling derivation information of moving as merging time domain candidate after the handling derivation information of moving.
12. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
Move from first order candidate and to select the derivation ensemble of communication adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving.
13. method according to claim 1 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
Said first order candidate whole candidates in derivation information derivation information of moving of moving is consolidated into a motion derivation information.
14. method according to claim 13 is characterized in that, said merging is handled, and comprising:
Do not have motion derivation information if said first order candidate moves in the derivation ensemble of communication, the motion derivation information after then said merging is handled is null vector;
Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an odd number if said first order candidate moves, and the motion derivation information after the then said merging processing is said spatial domain candidate the move vectorial intermediate value of derivation information of derivation information and time domain candidate of moving;
Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an even number if said first order candidate moves, and the motion derivation information after then said merging is handled is said spatial domain candidate derivation information and the time domain candidate derivation information mean value of two the motion derivation information in centre after the sequence arrangement by size that moves that moves.
15., it is characterized in that according to the described method of arbitrary claim in the claim 1～14, saidly said current block is carried out encoding process according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving, comprising:
If moving, said second level candidate comprises two candidates derivation information of moving in the derivation ensemble of communication; Then, said second level candidate selects a motion derivation information from moving the derivation ensemble of communication; Said current block is carried out encoding process; Obtain the encoding code stream of said current block, comprise the index coding of motion derivation information in said second level candidate moves the derivation ensemble of communication of selection in the said encoding code stream;
Comprise the candidate derivation information of moving in the derivation ensemble of communication if said second level candidate moves, then said current block is carried out encoding process, do not comprise the index coding in the encoding code stream of said current block.
16. method according to claim 15 is characterized in that, the index of motion derivation information in said second level candidate moves the derivation ensemble of communication of selecting carried out encoding process, comprising:
If the motion derivation information of selecting is time domain motion derivation information, then the index of said time domain motion derivation information in said second level candidate moves the derivation ensemble of communication is encoded to first coding; If the motion derivation information of selecting is spatial domain motion derivation information, then the index of said spatial domain motion derivation information in said second level candidate moves the derivation ensemble of communication is encoded to second coding.
17., it is characterized in that said motion derivation information is the MV under MVP or the Merge pattern according to the described method of arbitrary claim in the claim 1～14.
18. a video coding processing method is characterized in that, comprising:
The candidate who the obtains current block derivation ensemble of communication of moving;
There is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, then said current block carried out encoding process, obtain the encoding code stream of said current block, do not comprise motion derivation information index coding in the said encoding code stream.
19. method according to claim 18 is characterized in that, said method also comprises, has time domain motion derivation information and spatial domain motion derivation information if said candidate moves in the derivation ensemble of communication, then:
, said candidate selects a motion derivation information from moving the derivation ensemble of communication;
The motion derivation information of selecting is carried out block code handle, the number of derivation information is irrelevant so that encoding code stream and said candidate move;
The index of motion derivation information in said candidate moves the derivation ensemble of communication to selecting carries out encoding process, and said current block is carried out encoding process, obtains the encoding code stream of said current block, comprises the index coding in the said encoding code stream.
20., it is characterized in that said motion derivation information is the MV under MVP or the Merge pattern according to claim 18 or 19 described methods.
21. a video decode processing method is characterized in that, comprising:
Receive the encoding code stream of current block;
The first order candidate who the obtains current block derivation ensemble of communication of moving;
Candidate in the derivation ensemble of communication moves the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
According to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving the encoding code stream of said current block is carried out decoding processing.
22. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving.
23. method according to claim 22 is characterized in that, said merging is handled, and comprising:
Move whole spatial domain candidates in the derivation ensemble of communication of said first order candidate are moved the mean value of derivation information as merging spatial domain candidate after the handling derivation information of moving.
24. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 2; Then the said spatial domain candidate derivation information of moving is consolidated at the most two spatial domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the spatial domain candidates at the most that merge after handling derivation information of moving.
25. method according to claim 24 is characterized in that, said merging is handled and is comprised:
According to preset strategy, move from said first order candidate and to select the derivation ensemble of communication at the most two spatial domain candidates to move derivation information as merging spatial domain candidate after the handling derivation information of moving.
26. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate do not exist in the derivation ensemble of communication move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving.
27. method according to claim 26 is characterized in that, said merging is handled, and comprising:
According to preset strategy, move from said first order candidate and to select the derivation ensemble of communication at the most two time domain candidates to move derivation information as merging time domain candidate after the handling derivation information of moving.
28. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
If moving, said first order candidate exist in the derivation ensemble of communication move derivation information and time domain candidate of a spatial domain candidate to move the number of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises a time domain candidate who merges after handling derivation information and the said spatial domain candidate derivation information of moving of moving.
29. method according to claim 28 is characterized in that, said merging is handled, and comprising:
According to preset strategy, move from said first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information as merging time domain candidate after the handling derivation information of moving.
30. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
Move derivation information and time domain candidate of spatial domain candidate in the derivation ensemble of communication moves the individual number average of derivation information greater than 1 if said first order candidate moves; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling.
31. method according to claim 30 is characterized in that, said merging is handled, and comprising:
According to preset strategy; Move from said first order candidate and to select the derivation ensemble of communication time domain candidate to move derivation information, from said first order candidate moves the derivation ensemble of communication, select a spatial domain candidate to move derivation information as merging spatial domain candidate after the handling derivation information of moving as merging time domain candidate after the handling derivation information of moving.
32. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
Move from first order candidate and to select the derivation ensemble of communication adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving.
33. method according to claim 21 is characterized in that, said said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing, comprising:
Said first order candidate whole candidates in derivation information derivation information of moving of moving is consolidated into a motion derivation information.
34. method according to claim 33 is characterized in that, said merging is handled, and comprising:
Do not have motion derivation information if said first order candidate moves in the derivation ensemble of communication, the motion derivation information after then said merging is handled is null vector;
Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an odd number if said first order candidate moves, and the motion derivation information after the then said merging processing is said spatial domain candidate the move vectorial intermediate value of derivation information of derivation information and time domain candidate of moving;
Spatial domain candidate in the derivation ensemble of communication the move number sum of derivation information of derivation information and time domain candidate of moving is an even number if said first order candidate moves, and the motion derivation information after then said merging is handled is said spatial domain candidate derivation information and the time domain candidate derivation information mean value of two the motion derivation information in centre after the sequence arrangement by size that moves that moves.
35., it is characterized in that according to the described method of arbitrary claim in the claim 21～34, saidly the encoding code stream of said current block is carried out decoding processing according to the said second level candidate candidate that comprises in the derivation information derivation information of moving of moving, comprising:
If moving, said second level candidate comprises two candidates derivation information of moving in the derivation ensemble of communication; Then the motion derivation information index coding to said current block carries out decoding processing; Obtain index information; Move from said second level candidate according to said index information and to select motion derivation information the derivation ensemble of communication, and the encoding code stream of said current block is carried out decoding processing according to the motion derivation information of selecting;
Comprise the candidate derivation information of moving in the derivation information if said second level candidate moves, then the encoding code stream of said current block is carried out decoding processing according to this candidate derivation information of moving.
36., it is characterized in that said motion derivation information is the MV under MVP or the Merge pattern according to the described method of arbitrary claim in the claim 21～34.
37. a video decode processing method is characterized in that, comprising:
Receive the encoding code stream of current block;
The candidate who the obtains current block derivation ensemble of communication of moving;
There is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, then the encoding code stream of said current block carried out decoding processing according to the non-spatial domain candidate derivation information of moving.
38., it is characterized in that said method also comprises, has time domain motion derivation information and spatial domain motion derivation information if said candidate moves in the derivation information according to the described method of claim 37, then:
First to motion derivation information index coding is carried out decoding processing, obtains first information;
If corresponding first index of said first information, then confirm the said candidate motion derivation information in the derivation information of moving according to said first index;
If corresponding second index of said first information; Then the next bit to said index coding carries out decoding processing; Obtain next bit information; If said first information and next bit information correspondence second index, then confirm the said candidate motion derivation information in the derivation information of moving according to said second index;
And the like, be worth shared bit number up to motion derivation information index and reach maximum.
39. a code device is characterized in that, comprising:
First acquisition module, the first order candidate who is used to the to obtain current block derivation ensemble of communication of moving;
First merges processing module; Be used for the candidate of derivation ensemble of communication and move the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
The first encoding process module is used for candidate that the derivation ensemble of communication the comprises derivation information of moving of moving according to said second level candidate, and said current block is carried out encoding process.
40. according to the described device of claim 39; It is characterized in that; The said first derivation ensemble of communication that merges that processing module specifically is used for if said first order candidate moves exists move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving; Perhaps,
Specifically being used for if said first order candidate moves derivation ensemble of communication does not exist move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 2; Then the said spatial domain candidate derivation information of moving is consolidated at the most two spatial domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the spatial domain candidates at the most that merge after handling derivation information of moving; Perhaps,
Specifically being used for if said first order candidate moves derivation ensemble of communication does not exist move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving; Perhaps,
Specifically being used for if said first order candidate moves derivation ensemble of communication exists move derivation information and time domain candidate of a spatial domain candidate to move the number of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises a time domain candidate who merges after handling derivation information and the said spatial domain candidate derivation information of moving of moving; Perhaps,
Specifically be used for move derivation information and time domain candidate of if said first order candidate the moves spatial domain candidate of derivation ensemble of communication and move the individual number average of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling; Perhaps,
Specifically be used for selecting adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving from the first order candidate derivation ensemble of communication of moving; Perhaps,
Specifically be used for said first order candidate whole candidates of derivation information derivation information of moving of moving is consolidated into a motion derivation information.
41. according to claim 39 or 40 described devices; It is characterized in that; The said first encoding process module is used for specifically that the derivation ensemble of communication comprises two candidates derivation information of moving if said second level candidate moves; Then, said second level candidate selects a motion derivation information from moving the derivation ensemble of communication; Said current block is carried out encoding process, obtain the encoding code stream of said current block, comprise the index coding of motion derivation information in said second level candidate moves the derivation ensemble of communication of selection in the said encoding code stream; Comprise the candidate derivation information of moving in the derivation ensemble of communication if said second level candidate moves, then said current block is carried out encoding process, do not comprise the index coding in the encoding code stream of said current block.
42. a code device is characterized in that, comprising:
Second acquisition module, the candidate who is used to the to obtain current block derivation ensemble of communication of moving;
The second encoding process module; Be used for that there is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves; Then said current block is carried out encoding process, obtain the encoding code stream of said current block, do not comprise motion derivation information index coding in the said encoding code stream.
43. a decoding device is characterized in that, comprising:
First receiver module is used to receive the encoding code stream of current block;
The 3rd acquisition module, the first order candidate who is used to the to obtain said current block derivation ensemble of communication of moving;
Second merges processing module; Be used for the candidate of derivation ensemble of communication and move the number of derivation information greater than 2 if said first order candidate moves; Then said first order candidate candidate in the derivation ensemble of communication derivation information of moving of moving is merged processing; Obtain the second level candidate derivation ensemble of communication of moving, the move candidate that comprises in the derivation ensemble of communication of said second level candidate moves the number of derivation information smaller or equal to 2;
First decoding process module carries out decoding processing according to the said second level candidate candidate that comprises in the derivation ensemble of communication derivation information of moving of moving to the encoding code stream of said current block.
44. according to the described device of claim 43; It is characterized in that; The said second derivation ensemble of communication that merges that processing module specifically is used for if said first order candidate moves exists move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 1; Then the said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises the spatial domain candidate who merges after handling derivation information and the said time domain candidate derivation information of moving of moving; Perhaps,
Specifically being used for if said first order candidate moves derivation ensemble of communication does not exist move derivation information and spatial domain candidate of time domain candidate to move the number of derivation information greater than 2; Then the said spatial domain candidate derivation information of moving is consolidated at the most two spatial domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the spatial domain candidates at the most that merge after handling derivation information of moving; Perhaps,
Specifically being used for if said first order candidate moves derivation ensemble of communication does not exist move derivation information and time domain candidate of spatial domain candidate to move the number of derivation information greater than 2; Then the said time domain candidate derivation information of moving is consolidated at the most two time domain candidates derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises two the time domain candidates at the most that merge after handling derivation information of moving; Perhaps,
Specifically being used for if said first order candidate moves derivation ensemble of communication exists move derivation information and time domain candidate of a spatial domain candidate to move the number of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving, the said second level candidate derivation ensemble of communication of moving comprises a time domain candidate who merges after handling derivation information and the said spatial domain candidate derivation information of moving of moving; Perhaps,
Specifically be used for move derivation information and time domain candidate of if said first order candidate the moves spatial domain candidate of derivation ensemble of communication and move the individual number average of derivation information greater than 1; Then the said time domain candidate derivation information of moving is consolidated into the time domain candidate derivation information of moving; The said spatial domain candidate derivation information of moving is consolidated into the spatial domain candidate derivation information of moving, and the said second level candidate derivation ensemble of communication of moving comprises move derivation information and merge a spatial domain candidate after the handling derivation information of moving of a time domain candidate who merges after handling; Perhaps,
Specifically be used for selecting adjacent motion derivation information on the adjacent motion derivation information in a left side, spatial domain and the spatial domain as merging candidate after the processing derivation information of moving from the first order candidate derivation ensemble of communication of moving; Perhaps,
Specifically be used for said first order candidate whole candidates of derivation information derivation information of moving of moving is consolidated into a motion derivation information.
45. according to claim 43 or 44 described devices; It is characterized in that; Said first decoding process module is used for specifically that the derivation ensemble of communication comprises two candidates derivation information of moving if said second level candidate moves; Then the motion derivation information index coding to said current block carries out decoding processing; Obtain index information, move from said second level candidate according to said index information and select motion derivation information the derivation ensemble of communication, and the encoding code stream of said current block is carried out decoding processing according to the motion derivation information of selecting; Comprise the candidate derivation information of moving in the derivation information if said second level candidate moves, then the encoding code stream of said current block is carried out decoding processing according to this candidate derivation information of moving.
46. a decoding device is characterized in that, comprising:
Second receiver module is used to receive the encoding code stream of current block;
The 4th acquisition module, the candidate who is used to the to obtain current block derivation ensemble of communication of moving;
Second decoding process module is used for that there is not the spatial domain candidate derivation information of moving in the derivation ensemble of communication if said candidate moves, and then according to the non-spatial domain candidate derivation information of moving the encoding code stream of said current block is carried out decoding processing.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100600562A CN102685479A (en) | 2011-03-11 | 2011-03-11 | Video encoding and decoding processing method and device |
| PCT/CN2011/083831 WO2012122836A1 (en) | 2011-03-11 | 2011-12-12 | Video encoding/decoding method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100600562A CN102685479A (en) | 2011-03-11 | 2011-03-11 | Video encoding and decoding processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102685479A true CN102685479A (en) | 2012-09-19 |
Family
ID=46816770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100600562A Pending CN102685479A (en) | 2011-03-11 | 2011-03-11 | Video encoding and decoding processing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102685479A (en) |
| WO (1) | WO2012122836A1 (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104704835A (en) * | 2012-10-03 | 2015-06-10 | 联发科技股份有限公司 | Method and apparatus of motion information management in video coding |
| CN105325002A (en) * | 2013-04-23 | 2016-02-10 | 三星电子株式会社 | Multi-viewpoint video encoding method using viewpoint synthesis prediction and apparatus for same, and multi-viewpoint video decoding method and apparatus for same |
| CN110719463A (en) * | 2018-07-14 | 2020-01-21 | 北京字节跳动网络技术有限公司 | Extending look-up table based motion vector prediction with temporal information |
| CN112997480A (en) * | 2018-11-10 | 2021-06-18 | 北京字节跳动网络技术有限公司 | Rounding in pairwise averaging candidate computation |
| US11968377B2 (en) | 2018-06-21 | 2024-04-23 | Beijing Bytedance Network Technology Co., Ltd | Unified constrains for the merge affine mode and the non-merge affine mode |
| US11973962B2 (en) | 2018-06-05 | 2024-04-30 | Beijing Bytedance Network Technology Co., Ltd | Interaction between IBC and affine |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1518833A (en) * | 2002-04-19 | 2004-08-04 | ���µ�����ҵ��ʽ���� | Method for calculation motion vector |
| WO2009112742A1 (en) * | 2008-02-21 | 2009-09-17 | France Telecom | Encoding and decoding of an image or image sequence divided into pixel blocks |
| CN101860754A (en) * | 2009-12-16 | 2010-10-13 | 香港应用科技研究院有限公司 | Method and device for coding and decoding motion vector |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| BRPI0307119B1 (en) * | 2002-11-25 | 2018-02-14 | Godo Kaisha Ip Bridge 1 | “Mobile Image Decoding Method and Appliance” |
| CN101072356B (en) * | 2006-05-12 | 2011-02-09 | 中国科学院计算技术研究所 | Motion vector predicating method |
-
2011
- 2011-03-11 CN CN2011100600562A patent/CN102685479A/en active Pending
- 2011-12-12 WO PCT/CN2011/083831 patent/WO2012122836A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1518833A (en) * | 2002-04-19 | 2004-08-04 | ���µ�����ҵ��ʽ���� | Method for calculation motion vector |
| WO2009112742A1 (en) * | 2008-02-21 | 2009-09-17 | France Telecom | Encoding and decoding of an image or image sequence divided into pixel blocks |
| CN101860754A (en) * | 2009-12-16 | 2010-10-13 | 香港应用科技研究院有限公司 | Method and device for coding and decoding motion vector |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104704835A (en) * | 2012-10-03 | 2015-06-10 | 联发科技股份有限公司 | Method and apparatus of motion information management in video coding |
| CN104704835B (en) * | 2012-10-03 | 2017-11-24 | 联发科技股份有限公司 | The apparatus and method of movable information management in Video coding |
| CN105325002A (en) * | 2013-04-23 | 2016-02-10 | 三星电子株式会社 | Multi-viewpoint video encoding method using viewpoint synthesis prediction and apparatus for same, and multi-viewpoint video decoding method and apparatus for same |
| US11973962B2 (en) | 2018-06-05 | 2024-04-30 | Beijing Bytedance Network Technology Co., Ltd | Interaction between IBC and affine |
| US11968377B2 (en) | 2018-06-21 | 2024-04-23 | Beijing Bytedance Network Technology Co., Ltd | Unified constrains for the merge affine mode and the non-merge affine mode |
| CN110719463A (en) * | 2018-07-14 | 2020-01-21 | 北京字节跳动网络技术有限公司 | Extending look-up table based motion vector prediction with temporal information |
| CN110719463B (en) * | 2018-07-14 | 2022-11-22 | 北京字节跳动网络技术有限公司 | Extending look-up table based motion vector prediction with temporal information |
| CN112997480A (en) * | 2018-11-10 | 2021-06-18 | 北京字节跳动网络技术有限公司 | Rounding in pairwise averaging candidate computation |
| CN112997480B (en) * | 2018-11-10 | 2023-08-22 | 北京字节跳动网络技术有限公司 | Rounding in paired average candidate calculation |
| US11792421B2 (en) | 2018-11-10 | 2023-10-17 | Beijing Bytedance Network Technology Co., Ltd | Rounding in pairwise average candidate calculations |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2012122836A1 (en) | 2012-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110809161B (en) | Motion information candidate list construction method and device | |
| KR102540190B1 (en) | Intra prediction-based image coding method and apparatus using mpm list | |
| CN101478672B (en) | Video encoding, decoding method and apparatus, video processing system | |
| CN102656890B (en) | For the method and apparatus of Video coding | |
| CN103220508B (en) | Coding and decoding method and device | |
| CN104768011A (en) | Image encoding and decoding method and related device | |
| CN104769947B (en) | A kind of more hypothesis motion compensation encoding methods based on P frame | |
| CN103891290A (en) | Motion vector processing | |
| CN109587479A (en) | Inter-frame prediction method, device and the codec of video image | |
| CN102860006A (en) | Managing predicted motion vector candidates | |
| CN101415115A (en) | Method for encoding and decoding video based on movement dancing mode, and encoder and decoder thereof | |
| CN104243982A (en) | Video coding and decoding processing method and device | |
| CN103079067B (en) | Motion vector predictor list builder method and video coding-decoding method and device | |
| CN109922336B (en) | Inter-frame prediction method and device for video data | |
| CN102685479A (en) | Video encoding and decoding processing method and device | |
| TW202021354A (en) | Motion vector predictor list generation | |
| KR20140029383A (en) | Image coding device and image decoding device | |
| JP7223111B2 (en) | Motion vector prediction method and apparatus, encoder and decoder | |
| KR20170069917A (en) | Method and apparatus for encoding and decoding an information indicating a prediction method of an intra skip mode | |
| CN106604028A (en) | Encoding processing method and device, decoding processing method and device, encoder, and decoder | |
| JP7448558B2 (en) | Methods and devices for image encoding and decoding | |
| KR20200005575A (en) | Method and apparatus for determining motion vector | |
| CN113709458A (en) | Displacement vector prediction method, device and equipment in video coding and decoding | |
| CN113678455B (en) | Video or image coding for deriving weight index information for bi-prediction | |
| CN111669592B (en) | Encoding and decoding method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120919 |