CN104902271B - Predicting mode selecting method and device - Google Patents
Predicting mode selecting method and device Download PDFInfo
- Publication number
- CN104902271B CN104902271B CN201510249916.5A CN201510249916A CN104902271B CN 104902271 B CN104902271 B CN 104902271B CN 201510249916 A CN201510249916 A CN 201510249916A CN 104902271 B CN104902271 B CN 104902271B
- Authority
- CN
- China
- Prior art keywords
- rate distortion
- patterns
- distortion costs
- interframe
- merge
- 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.)
- Active
Links
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of predicting mode selecting method and device, belong to field of video encoding.This method includes：Rate distortion costs J of the current prediction unit under using interframe skip patterns, interframe merge patterns and interframe 2N*2N patterns is calculated respectivelyskip、JmergeAnd J2N*2N；Detect JskipWhether J is less thanmergeAnd J2N*2N；If JskipLess than JmergeAnd J2N*2N, then by JskipIt is arranged to candidate's rate distortion costs Jmode；Calculate rate distortion costs J of current prediction unit when using at least one intra prediction modeintra；From JmodeWith at least one JintraIn select the minimum predictive mode of rate distortion costs.The present invention skips the rate distortion costs calculating process of other 7 kinds of inter-frame forecast modes when interframe skip patterns are more excellent, can effectively reduce the time-consuming of predictive mode selection.
Description
Technical field
The present invention relates to field of video encoding, more particularly to a kind of predicting mode selecting method and device.
Background technology
In HEVC (High Efficient Video Coding, high efficiency Video coding) standard, it is proposed that prediction is single
Member (Prediction Unit, referred to as：PU concept).Predicting unit is to carry out the elementary cell of inter prediction and infra-frame prediction.
In HEVC standard, there is provided several inter-frame forecast mode and several intra prediction mode are available.For
Optimal predictive mode is found, prior art needs to calculate rate distortion of the current prediction unit under each predictive mode
Cost, and the minimum predictive mode of rate distortion costs is selected as the predictive mode finally used.
During the present invention is realized, inventor has found that prior art at least has problems with：Due to needing to calculate
Rate distortion costs under each predictive mode, and the amount of calculation of rate distortion costs is larger, causes the selection course of predictive mode
Need to spend it is more time-consuming, account for whole cataloged procedure it is time-consuming 60%~70%.
The content of the invention
In order to solve problem of the prior art, the embodiments of the invention provide a kind of predicting mode selecting method and device.
The technical scheme is as follows：
According to the first aspect of the disclosure, there is provided a kind of predicting mode selecting method, methods described include：
Rate distortion costs J of current prediction unit when using interframe skip patterns is calculated respectivelyskip, using interframe
Rate distortion costs J during merge patternsmergeWith the rate distortion costs J under use interframe 2N*2N patterns2N*2N；
Detect the JskipWhether the J is less thanmergeWith the J2N*2N；
If the JskipLess than the JmergeWith the J2N*2N, then by the JskipIt is arranged to candidate's rate distortion costs
Jmode；
Calculate rate distortion costs J of current prediction unit when using at least one intra prediction modeintra；
From the JmodeWith at least one JintraIn select described in the minimum predictive mode of rate distortion costs is used as
The final predictive mode of current prediction unit.
According to the second aspect of the disclosure, there is provided a kind of prediction mode selection apparatus, described device include：
First computing module, for calculating rate distortion costs of current prediction unit when using interframe skip patterns respectively
Jskip, using rate distortion costs J during interframe merge patternsmergeWith the rate distortion costs under use interframe 2N*2N patterns
J2N*2N；
Skip detection module, for detecting the JskipWhether the J is less thanmergeWith the J2N*2N；
Setup module is skipped, if for the JskipLess than the JmergeWith the J2N*2N, then by the JskipIt is arranged to
Candidate's rate distortion costs Jmode；
Second computing module, it is additionally operable to calculate the current prediction unit when using at least one intra prediction mode
Rate distortion costs Jintra；
Final decision module, for from the JmodeWith at least one JintraIn to select rate distortion costs minimum
Final predictive mode of the predictive mode as the current prediction unit.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought is：
By when interframe skip patterns are more excellent, skipping except interframe skip patterns, interframe merge patterns and interframe 2N*2N
The rate distortion costs calculating process of other 7 kinds of inter-frame forecast modes outside pattern, reduces needed for predictive mode selection course
The amount of calculation wanted, the time-consuming of predictive mode selection can be effectively reduced, meets the demand of the higher scene of some requirement of real-time.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 be the present embodiments relate to predicting unit 8 kinds of inter-frame forecast modes division schematic diagram；
Fig. 2 is the method flow diagram for the predicting mode selecting method that one embodiment of the invention provides；
Fig. 3 is the method flow diagram for the predicting mode selecting method that another embodiment of the present invention provides；
Fig. 4 is the structural representation for the prediction mode selection apparatus that one embodiment of the invention provides；
Fig. 5 is the structural representation for the prediction mode selection apparatus that another embodiment of the present invention provides.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is described in further detail.
Brief introduction is carried out to several nouns involved by the present embodiment first：
Picture frame：One section of video is made up of some two field picture frames.Video compression coding is typically taken based on the coding staff of block
Formula, i.e., the two field picture frame in video is divided into the block of multiple non-overlapping copies, these blocks encoded afterwards.Each picture frame
Prediction interframe encoding mode or intraframe predictive coding mode can be used in Video coding.
In HEVC standard, it is proposed that the concept of coding unit, predicting unit and converter unit.
Coding unit：It is the elementary cell encoded in a picture frame.Coding unit can be 64*64 pixel
Block.
Predicting unit：It is the elementary cell being predicted in a picture frame.Predicting unit can be 64*64,32*32,
The block of the Pixel Dimensions such as 16*16,8*8.
Converter unit：It is that residual error (Residual) or conversion coefficient (Transform are presented in a picture frame
Coefficients elementary cell).Converter unit can also be the block of the Pixel Dimensions such as 32*32,16*16,8*8,4*4.This
It is not related to the discussion of coding unit and converter unit in text.
For a predicting unit, 10 kinds of inter-frame forecast modes and 3 kinds of intra prediction modes are provided in HEVC standard.
10 kinds of inter-frame forecast modes include：Interframe skip patterns, interframe merge patterns, interframe 2N*2N patterns, interframe N*N
Pattern, interframe 2N*N patterns, interframe N*2N patterns, interframe 2N*UD patterns, interframe 2N*nD patterns, interframe nL*2N patterns, interframe
NR*2N patterns, as shown in Figure 1.
Wherein, interframe 2N*2N patterns, interframe N*N patterns, interframe 2N*N patterns, interframe N*2N patterns belong to four kinds symmetrically
Prediction mode；Interframe 2N*UD patterns and interframe 2N*nD patterns belong to the asymmetric prediction mode of horizontal direction；Interframe nL*2N moulds
Formula, interframe nR*2N patterns belong to the asymmetric prediction mode of vertical direction.
3 kinds of intra prediction modes include：2N*2N patterns in frame, PCM patterns in N*N patterns and frame in frame.
In HEVC standard, when the optimal prediction modes of decision-making current prediction unit, it is necessary to calculate current prediction unit
Using rate distortion costs corresponding to difference during above-mentioned 13 kinds of predictive modes, amount of calculation is very big, causes the selection of predictive mode
Journey need to spend it is more time-consuming, account for whole cataloged procedure it is time-consuming 60%~70%.
Fig. 2 is refer to, the method flow diagram of the predicting mode selecting method provided it illustrates one embodiment of the invention.
The present embodiment is applied to illustrate in video encoder with the predicting mode selecting method.This method includes：
Step 201, rate distortion costs J of current prediction unit when using interframe skip patterns is calculated respectivelyskip, use
Rate distortion costs J during interframe merge patternsmergeWith the rate distortion costs J under use interframe 2N*2N patterns2N*2N；
Rate distortion costs, it is the letter of rate-distortion optimization (Rate-distortion optimization, abbreviation RDO) cost
Claim.The calculating process of rate distortion costs is prior art, is repeated no more herein.
Step 202, J is detectedskipWhether J is less thanmergeAnd J2N*2N；
Step 203, if JskipLess than JmergeAnd J2N*2N, then by JskipIt is arranged to candidate's rate distortion costs Jmode；
Step 204, rate distortion costs J of current prediction unit when using at least one intra prediction mode is calculatedintra；
Step 205, from JmodeWith at least one JintraIn select the minimum predictive mode of rate distortion costs as current
The final predictive mode of predicting unit.
In summary, the predicting mode selecting method that the present embodiment provides, when interframe skip patterns are more excellent, is skipped except frame
Between other 7 kinds of inter-frame forecast modes outside skip patterns, interframe merge patterns and interframe 2N*2N patterns rate distortion costs
Calculating process, reduce the amount of calculation required for predictive mode selection course, can effectively reduce the time-consuming of predictive mode selection,
Meet the demand of the higher scene of some requirement of real-time.
Fig. 3 is refer to, the method flow of the predicting mode selecting method provided it illustrates another embodiment of the present invention
Figure.The present embodiment is applied to illustrate in video encoder with the predicting mode selecting method.This method includes：
Step 301, rate distortion costs J of current prediction unit when using interframe skip patterns is calculated respectivelyskip, use
Rate distortion costs J during interframe merge patternsmergeWith the rate distortion costs J under use interframe 2N*2N patterns2N*2N；
Step 302, J is detectedskipWhether J is less thanmergeAnd J2N*2N；
If JskipLess than JmergeAnd J2N*2N, then into step 303；
If JskipNot less than JmergeAnd J2N*2N, then into step 304.
Step 303, by JskipIt is arranged to candidate's rate distortion costs Jmode；
Step 304, rate distortion costs J of current prediction unit when using interframe N*N patterns is calculated respectivelyN*N, using frame
Between 2N*N patterns when rate distortion costs J2N*NWith the rate distortion costs J under use interframe N*2N patternsN*2N。
Step 305, from Jmerge、J2N*2N、JN*N、J2N*NAnd JN*2NIn find out the first minimum value；
If the first minimum value is Jmerge, then into step 306.
If the first minimum value is not Jmerge, then into step 307.
Step 306, the first minimum value is arranged to candidate's rate distortion costs Jmode；
Step 307, detect whether to open the asymmetric predictive mode of horizontal direction and/or vertical direction；
Whether the asymmetric predictive mode of horizontal direction and/or vertical direction is opened in video encoder, by outside
Coding parameter is preset.
If being both not turned on the asymmetric predictive mode of horizontal direction, the asymmetric prediction mould of vertical direction is also not turned on
Formula, then into step 306.
If the asymmetric predictive mode of horizontal direction is only opened, into step 308；
If the asymmetric predictive mode of vertical direction is only opened, into step 310；
If asymmetric predictive mode horizontally and vertically is opened simultaneously, into step 316.
Step 308, rate distortion costs J of current prediction unit when using interframe 2N*nU patterns is calculated respectively2N*nU, adopt
With rate distortion costs J during interframe 2N*nD patterns2N*nD；
Step 309, from the first minimum value, J2N*nUAnd J2N*nDIn find out the second minimum value；
Step 310, rate distortion costs J of current prediction unit when using interframe nL*2N patterns is calculated respectivelynL*2N, adopt
With rate distortion costs J during interframe nR*2N patternsnR*2N；
Step 311, from the first minimum value, JnL*2NAnd JnR*2NIn find out the second minimum value；
Step 312, the second minimum value is arranged to candidate's rate distortion costs Jmode；
Step 313, rate distortion costs J of current prediction unit when using interframe 2N*nU patterns is calculated respectively2N*nU, adopt
With rate distortion costs J during interframe 2N*nD patterns2N*nD, using rate distortion costs J during interframe nL*2N patternsnL*2N, using frame
Between nR*2N patterns when rate distortion costs JnR*2N；
Step 314, from the first minimum value, J2N*nU、J2N*nD、JnL*2NAnd JnR*2NIn find out the 3rd minimum value；
Step 315, the 3rd minimum value is arranged to candidate's rate distortion costs Jmode；
Step 316, rate distortion costs of current prediction unit when using 2N*2N patterns in frame are calculated respectively
Jintra_2N*2N, using the rate distortion costs J in frame during N*N patternsintra_N*N, using the rate distortion costs in frame during PCM patterns
Jintra_PCM；
Intra prediction mode includes 2N*2N patterns in frame, at least one of PCM patterns in N*N patterns and frame in frame.
Step 317, in Jmode、Jintra_2N*2N、Jintra_N*N、Jintra_PCMIn select the 4th minimum value, it is minimum by the 4th
Final predictive mode of the predictive mode corresponding to value as predicting unit.
In summary, the predicting mode selecting method that the present embodiment provides, when interframe skip patterns are more excellent, is skipped except frame
Between other 7 kinds of inter-frame forecast modes outside skip patterns, interframe merge patterns and interframe 2N*2N patterns rate distortion costs
Calculating process, reduce the amount of calculation required for predictive mode selection course, can effectively reduce the time-consuming of predictive mode selection,
Meet the demand of the higher scene of some requirement of real-time.
When interframe merge patterns are more excellent, even if opening the asymmetric prediction mould of horizontal direction and/or vertical direction
Formula, the rate distortion costs calculating process of 4 kinds of asymmetric predictive modes about horizontal direction and/or vertical direction is also skipped, is subtracted
The amount of calculation required for predictive mode selection course is lacked, can effectively reduce the time-consuming of predictive mode selection, meet some realities
When property requires the demand of higher scene.
The predicting mode selecting method that the present embodiment provides is being done greatly corresponding to HEVC coding standards in reference software
Amount experiment, on the premise of Image Coding quality is ensured, coding rate can be made averagely to improve 42% or so, code efficiency loss control
System is within 0.5%.
It should be noted that simultaneously to whether enabling asymmetric prediction horizontally and vertically in step 306
Pattern is detected.In other embodiments, can also be first to whether enabling the asymmetric predictive mode of horizontal direction and entering
Row detection, then again to whether enabling the asymmetric predictive mode of vertical direction and detecting；Or can also first to whether
The asymmetric predictive mode for enabling vertical direction is detected, then again to whether enabling the asymmetric prediction of horizontal direction
Pattern is detected.The present embodiment is not specifically limited to this.
It is below the device embodiment of the present invention, the details not being described in detail in device embodiment, it is above-mentioned right to may be referred to
The embodiment of the method answered.
Fig. 4 is refer to, the block diagram of the prediction mode selection apparatus provided it illustrates one embodiment of the invention.
The prediction mode selection apparatus can by software, hardware or both be implemented in combination with as video encoder whole or
A part.The prediction mode selection apparatus, including：
First computing module 410, for calculating rate distortion of current prediction unit when using interframe skip patterns respectively
Cost Jskip, using rate distortion costs J during interframe merge patternsmergeWith the rate distortion costs under use interframe 2N*2N patterns
J2N*2N；
Skip detection module 420, for detecting the JskipWhether the J is less thanmergeWith the J2N*2N；
Setup module 430 is skipped, if for the JskipLess than the JmergeWith the J2N*2N, then by the JskipSet
For candidate's rate distortion costs Jmode；
Second computing module 440, it is additionally operable to calculate the current prediction unit at least one intra prediction mode
Rate distortion costs Jintra；
Final decision module 450, for from the JmodeWith at least one JintraIn select rate distortion costs most
Final predictive mode of the small predictive mode as the current prediction unit.
In summary, the prediction mode selection apparatus that the present embodiment provides, when interframe skip patterns are more excellent, is skipped except frame
Between other 7 kinds of inter-frame forecast modes outside skip patterns, interframe merge patterns and interframe 2N*2N patterns rate distortion costs
Calculating process, reduce the amount of calculation required for predictive mode selection course, can effectively reduce the time-consuming of predictive mode selection,
Meet the demand of the higher scene of some requirement of real-time.
Fig. 5 is refer to, the block diagram of the prediction mode selection apparatus provided it illustrates one embodiment of the invention.
The prediction mode selection apparatus can by software, hardware or both be implemented in combination with as video encoder whole or
A part.The prediction mode selection apparatus, including：
First computing module 410, for calculating rate distortion of current prediction unit when using interframe skip patterns respectively
Cost Jskip, using rate distortion costs J during interframe merge patternsmergeWith the rate distortion costs under use interframe 2N*2N patterns
J2N*2N；
Skip detection module 420, for detecting the JskipWhether the J is less thanmergeWith the J2N*2N；
Setup module 430 is skipped, if for the JskipLess than the JmergeWith the J2N*2N, then by the JskipSet
For candidate's rate distortion costs Jmode；
Second computing module 440, it is additionally operable to calculate the current prediction unit at least one intra prediction mode
Rate distortion costs Jintra；
Final decision module 450, for from the JmodeWith at least one JintraIn select rate distortion costs most
Final predictive mode of the small predictive mode as the current prediction unit.
Alternatively, described device, in addition to：
3rd computing module 462, if being additionally operable to the JskipNot less than the JmergeWith the J2N*2N, then institute is calculated respectively
State rate distortion costs J of current prediction unit when using interframe N*N patternsN*N, using rate distortion generation during interframe 2N*N patterns
Valency J2N*NWith the rate distortion costs J under use interframe N*2N patternsN*2N；
First searching modul 464, for from the Jmerge, the J2N*2N, the JN*N, the J2N*NWith the JN*2NIn
Find out the first minimum value；
First setup module 466, if being the J for first minimum valuemerge, then by the JmergeIt is arranged to institute
State candidate's rate distortion costs Jmode。
Alternatively, described device, in addition to：
Detection module 471 is enabled, if not being the J for first minimum valuemerge, then detect whether to open level
Direction and/or the asymmetric predictive mode of vertical direction；
4th computing module 473, if the asymmetric predictive mode for only opening horizontal direction, respectively described in calculating
Rate distortion costs J of current prediction unit when using interframe 2N*nU patterns2N*nU, using rate distortion during interframe 2N*nD patterns
Cost J2N*nD；
Second searching modul 475, for from first minimum value, the J2N*nUWith the J2N*nDIn find out second
Minimum value；
5th computing module 477, if the asymmetric predictive mode for only opening vertical direction, respectively described in calculating
Rate distortion costs J of current prediction unit when using interframe nL*2N patternsnL*2N, using rate distortion during interframe nR*2N patterns
Cost JnR*2N；
Second searching modul 475, it is additionally operable to from first minimum value, the JnL*2NWith the JnR*2NMiddle lookup
Go out second minimum value；
Second setup module 479, for second minimum value to be arranged into candidate's rate distortion costs Jmode。
Alternatively, described device, in addition to：
Detection module 471 is enabled, if not being the J for first minimum valuemerge, then the current predictive list is detected
Whether member opens the asymmetric predictive mode of horizontal direction and/or vertical direction；
6th computing module 472, if for opening asymmetric predictive mode horizontally and vertically simultaneously,
Rate distortion costs J of current prediction unit when using interframe 2N*nU patterns is calculated respectively2N*nU, using interframe 2N*nD moulds
Rate distortion costs J during formula2N*nD, using rate distortion costs J during interframe nL*2N patternsnL*2N, using interframe nR*2N patterns when
Rate distortion costs JnR*2N；
3rd searching modul 474, for from first minimum value, the J2N*nU, the J2N*nD, the JnL*2N
With the JnR*2NIn find out the 3rd minimum value；
3rd setup module 476, for the 3rd minimum value to be arranged into candidate's rate distortion costs Jmode。
Alternatively, second computing module 440, specifically for calculating the current prediction unit respectively using in frame
Rate distortion costs J during 2N*2N patternsintra_2N*2N, using the rate distortion costs J in frame during N*N patternsintra_N*N, using in frame
Rate distortion costs J during PCM patternsintra_PCM。
In summary, the prediction mode selection apparatus that the present embodiment provides, when interframe skip patterns are more excellent, is skipped except frame
Between other 7 kinds of inter-frame forecast modes outside skip patterns, interframe merge patterns and interframe 2N*2N patterns rate distortion costs
Calculating process, reduce the amount of calculation required for predictive mode selection course, can effectively reduce the time-consuming of predictive mode selection,
Meet the demand of the higher scene of some requirement of real-time.
The prediction mode selection apparatus that the present embodiment provides, also when interframe merge patterns are more excellent, even if opening level
Direction and/or the asymmetric predictive mode of vertical direction, also skip 4 kinds it is asymmetric about horizontal direction and/or vertical direction
The rate distortion costs calculating process of predictive mode, reduces the amount of calculation required for predictive mode selection course, can effectively drop
Taking for low predictive mode selection, meets the demand of the higher scene of some requirement of real-time.
It should be noted that：Above-described embodiment provide prediction mode selection apparatus in triggering intelligent network service, only with
The division progress of above-mentioned each functional module, can be as needed and by above-mentioned function distribution by not for example, in practical application
Same functional module is completed, i.e., the internal structure of equipment is divided into different functional modules, to complete whole described above
Or partial function.In addition, prediction mode selection apparatus and predicting mode selecting method embodiment category that above-described embodiment provides
In same design, its specific implementation process refers to embodiment of the method, repeated no more here.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, by program the hardware of correlation can also be instructed to complete, described program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (10)
1. a kind of predicting mode selecting method, it is characterised in that methods described includes：
Rate distortion costs J of current prediction unit when using interframe skip patterns is calculated respectivelyskip, using interframe merge patterns
When rate distortion costs JmergeWith the rate distortion costs J under use interframe 2N*2N patterns2N*2N；
Detect the JskipWhether the J is less thanmergeWith the J2N*2N；
If the JskipLess than the JmergeWith the J2N*2N, then by the JskipIt is arranged to candidate's rate distortion costs Jmode；
Calculate rate distortion costs J of current prediction unit when using at least one intra prediction modeintra；
From the JmodeWith at least one JintraIn select the minimum predictive mode of rate distortion costs as described current
The final predictive mode of predicting unit.
2. according to the method for claim 1, it is characterised in that the detection JskipWhether the J is less thanmergeAnd institute
State J2N*2NAfterwards, in addition to：
If the JskipNot less than the JmergeWith the J2N*2N, then the current prediction unit is calculated respectively is using interframe N*
Rate distortion costs J during N patternsN*N, using rate distortion costs J during interframe 2N*N patterns2N*NUnder use interframe N*2N patterns
Rate distortion costs JN*2N；
From the Jmerge, the J2N*2N, the JN*N, the J2N*NWith the JN*2NIn find out the first minimum value；
If first minimum value is the Jmerge, then by the JmergeIt is arranged to candidate's rate distortion costs Jmode。
3. according to the method for claim 2, it is characterised in that described from the Jmerge, the J2N*2N, the JN*N, it is described
J2N*NWith the JN*2NIn find out the first minimum value after, in addition to：
If first minimum value is not the Jmerge, then detect whether to open the non-right of horizontal direction and/or vertical direction
Claim predictive mode；
If only opening the asymmetric predictive mode of horizontal direction, the current prediction unit is calculated respectively using interframe
Rate distortion costs J during 2N*nU patterns2N*nU, using rate distortion costs J during interframe 2N*nD patterns2N*nD；From described first most
Small value, the J2N*nUWith the J2N*nDIn find out the second minimum value；
If only opening the asymmetric predictive mode of vertical direction, the current prediction unit is calculated respectively using interframe
Rate distortion costs J during nL*2N patternsnL*2N, using rate distortion costs J during interframe nR*2N patternsnR*2N；From described first most
Small value, the JnL*2NWith the JnR*2NIn find out second minimum value；
Second minimum value is arranged to candidate's rate distortion costs Jmode。
4. according to the method for claim 2, it is characterised in that described from the Jmerge, the J2N*2N, the JN*N, it is described
J2N*NWith the JN*2NIn find out the first minimum value after, in addition to：
If first minimum value is not the Jmerge, then detect the current prediction unit whether open horizontal direction and/
Or the asymmetric predictive mode of vertical direction；
If opening asymmetric predictive mode horizontally and vertically simultaneously, the current prediction unit is calculated respectively
Rate distortion costs J when using interframe 2N*nU patterns2N*nU, using rate distortion costs J during interframe 2N*nD patterns2N*nD, adopt
With rate distortion costs J during interframe nL*2N patternsnL*2N, using rate distortion costs J during interframe nR*2N patternsnR*2N；
From first minimum value, the J2N*nU, the J2N*nD, the JnL*2NWith the JnR*2NIn find out the 3rd minimum value；
3rd minimum value is arranged to candidate's rate distortion costs Jmode。
5. method according to any one of claims 1 to 4, it is characterised in that described to calculate the current prediction unit extremely
Rate distortion costs J during a kind of few intra prediction modeintra, including：
Rate distortion costs J of current prediction unit when using 2N*2N patterns in frame is calculated respectivelyintra_2N*2N, using frame
Rate distortion costs J during interior N*N patternsintra_N*N, using the rate distortion costs J in frame during PCM patternsintra_PCM。
6. a kind of prediction mode selection apparatus, it is characterised in that described device includes：
First computing module, for calculating rate distortion costs J of current prediction unit when using interframe skip patterns respectivelyskip、
Using rate distortion costs J during interframe merge patternsmergeWith the rate distortion costs J under use interframe 2N*2N patterns2N*2N；
Skip detection module, for detecting the JskipWhether the J is less thanmergeWith the J2N*2N；
Setup module is skipped, if for the JskipLess than the JmergeWith the J2N*2N, then by the JskipIt is arranged to candidate
Rate distortion costs Jmode；
Second computing module, it is additionally operable to calculate rate mistake of current prediction unit when using at least one intra prediction mode
True cost Jintra；
Final decision module, for from the JmodeWith at least one JintraIn select the minimum prediction of rate distortion costs
Final predictive mode of the pattern as the current prediction unit.
7. device according to claim 6, it is characterised in that described device, in addition to：
3rd computing module, if being additionally operable to the JskipNot less than the JmergeWith the J2N*2N, then calculate respectively described current
Rate distortion costs J of predicting unit when using interframe N*N patternsN*N, using rate distortion costs J during interframe 2N*N patterns2N*N
With the rate distortion costs J under use interframe N*2N patternsN*2N；
First searching modul, for from the Jmerge, the J2N*2N, the JN*N, the J2N*NWith the JN*2NIn find out
One minimum value；
First setup module, if being the J for first minimum valuemerge, then by the JmergeThe candidate is arranged to lead
Distortion cost Jmode。
8. device according to claim 7, it is characterised in that described device, in addition to：
Detection module is enabled, if not being the J for first minimum valuemerge, then detect whether to open horizontal direction and/
Or the asymmetric predictive mode of vertical direction；
4th computing module, if the asymmetric predictive mode for only opening horizontal direction, calculate respectively described current pre-
Survey rate distortion costs J of unit when using interframe 2N*nU patterns2N*nU, using rate distortion costs during interframe 2N*nD patterns
J2N*nD；
Second searching modul, for from first minimum value, the J2N*nUWith the J2N*nDIn find out the second minimum value；
5th computing module, if the asymmetric predictive mode for only opening vertical direction, calculate respectively described current pre-
Survey rate distortion costs J of unit when using interframe nL*2N patternsnL*2N, using rate distortion costs during interframe nR*2N patterns
JnR*2N；
Second searching modul, it is additionally operable to from first minimum value, the JnL*2NWith the JnR*2NIn find out described
Two minimum values；
Second setup module, for second minimum value to be arranged into candidate's rate distortion costs Jmode。
9. device according to claim 7, it is characterised in that described device, in addition to：
Detection module is enabled, if not being the J for first minimum valuemerge, then whether the current prediction unit is detected
Open the asymmetric predictive mode of horizontal direction and/or vertical direction；
6th computing module, if for opening asymmetric predictive mode horizontally and vertically simultaneously, count respectively
Calculate rate distortion costs J of current prediction unit when using interframe 2N*nU patterns2N*nU, using during interframe 2N*nD patterns
Rate distortion costs J2N*nD, using rate distortion costs J during interframe nL*2N patternsnL*2N, using during interframe nR*2N patterns rate lose
True cost JnR*2N；
3rd searching modul, for from first minimum value, the J2N*nU, the J2N*nD, the JnL*2NWith the JnR*2NIn
Find out the 3rd minimum value；
3rd setup module, for the 3rd minimum value to be arranged into candidate's rate distortion costs Jmode。
10. according to any described device of claim 6 to 9, it is characterised in that second computing module, specifically for dividing
Rate distortion costs J of current prediction unit when using 2N*2N patterns in frame is not calculatedintra_2N*2N, using N*N moulds in frame
Rate distortion costs J during formulaintra_N*N, using the rate distortion costs J in frame during PCM patternsintra_PCM。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510249916.5A CN104902271B (en) | 2015-05-15 | 2015-05-15 | Predicting mode selecting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510249916.5A CN104902271B (en) | 2015-05-15 | 2015-05-15 | Predicting mode selecting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104902271A CN104902271A (en) | 2015-09-09 |
| CN104902271B true CN104902271B (en) | 2017-12-29 |
Family
ID=54034631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510249916.5A Active CN104902271B (en) | 2015-05-15 | 2015-05-15 | Predicting mode selecting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104902271B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106534849B (en) * | 2016-12-30 | 2019-07-16 | 杭州当虹科技股份有限公司 | A kind of quick HEVC inter-frame encoding methods |
| CN108810531B (en) * | 2017-05-03 | 2019-11-19 | 腾讯科技（深圳）有限公司 | Video coding processing method, device and electronic equipment |
| CN108632615A (en) * | 2018-04-09 | 2018-10-09 | 首都师范大学 | A method of the HEVC based on motion vector analysis judges SKIP patterns in advance |
| CN110855998B (en) * | 2018-08-20 | 2023-04-11 | 华为技术有限公司 | Fusion candidate list construction method and device, and fusion candidate list editing/decoding method and device |
| WO2020038357A1 (en) * | 2018-08-20 | 2020-02-27 | 华为技术有限公司 | Fusion candidate list construction method, device and encoding/decoding method and device |
| CN108924551B (en) * | 2018-08-29 | 2022-01-07 | 腾讯科技（深圳）有限公司 | Method for predicting video image coding mode and related equipment |
| CN110149512B (en) * | 2018-09-14 | 2023-04-14 | 腾讯科技（深圳）有限公司 | Inter-frame prediction acceleration method, device, computer storage medium and equipment |
| CN111279699B (en) * | 2019-04-26 | 2024-06-14 | 深圳市大疆创新科技有限公司 | Video encoding and decoding method and device |
| CN112118450B (en) * | 2019-06-21 | 2022-03-29 | 杭州海康威视数字技术股份有限公司 | Method and device for decoding and encoding prediction mode |
| CN111277838B (en) * | 2020-02-17 | 2022-10-14 | 腾讯科技（深圳）有限公司 | Encoding mode selection method, device, electronic equipment and computer readable medium |
| CN111629206A (en) * | 2020-05-25 | 2020-09-04 | 腾讯科技（深圳）有限公司 | Video encoding method, video playing method, related device and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103155555A (en) * | 2010-10-01 | 2013-06-12 | 高通股份有限公司 | Coding prediction modes in video coding |
| CN103229507A (en) * | 2010-11-25 | 2013-07-31 | Lg电子株式会社 | Method for signaling image information, and method for decoding image information using same |
| CN104301739A (en) * | 2013-07-18 | 2015-01-21 | 联发科技（新加坡）私人有限公司 | Multi-view video coding method |
| CN104320656A (en) * | 2014-10-30 | 2015-01-28 | 上海交通大学 | Method for quickly selecting interframe encoding modes in x265 encoder |
| CN104333756A (en) * | 2014-11-19 | 2015-02-04 | 西安电子科技大学 | HEVC (High Efficiency Video Coding) prediction mode fast selection method based on time domain correlation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2495990A (en) * | 2011-10-28 | 2013-05-01 | Canon Kk | Motion compensated image coding with omission of coding mode cost calculation for a motion predictor in a set. |
-
2015
- 2015-05-15 CN CN201510249916.5A patent/CN104902271B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103155555A (en) * | 2010-10-01 | 2013-06-12 | 高通股份有限公司 | Coding prediction modes in video coding |
| CN103229507A (en) * | 2010-11-25 | 2013-07-31 | Lg电子株式会社 | Method for signaling image information, and method for decoding image information using same |
| CN104301739A (en) * | 2013-07-18 | 2015-01-21 | 联发科技（新加坡）私人有限公司 | Multi-view video coding method |
| CN104320656A (en) * | 2014-10-30 | 2015-01-28 | 上海交通大学 | Method for quickly selecting interframe encoding modes in x265 encoder |
| CN104333756A (en) * | 2014-11-19 | 2015-02-04 | 西安电子科技大学 | HEVC (High Efficiency Video Coding) prediction mode fast selection method based on time domain correlation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104902271A (en) | 2015-09-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104902271B (en) | Predicting mode selecting method and device | |
| CN104853192B (en) | Predicting mode selecting method and device | |
| EP3709648A1 (en) | Method and apparatus for video encoding and decoding with subblock based local illumination compensation | |
| CN103491369B (en) | A kind of interframe prediction encoding method and encoder | |
| CN105208387B (en) | A kind of HEVC Adaptive Mode Selection Method for Intra-Prediction | |
| CN100581255C (en) | Pixel loop filtering method and filter | |
| CN104639940B (en) | A kind of quick HEVC method for choosing frame inner forecast mode | |
| CN108347616A (en) | A kind of depth prediction approach and device based on optional time domain motion-vector prediction | |
| CN104811729B (en) | A kind of video multi-reference frame coding method | |
| CN103118262B (en) | Rate distortion optimization method and device, and video coding method and system | |
| CN107566846A (en) | Video coding skip mode decision-making technique, device, equipment and storage medium | |
| CN111988628B (en) | VVC rapid intra-frame coding method based on reinforcement learning | |
| CN102932642A (en) | Interframe coding quick mode selection method | |
| CN104883566B (en) | The fast algorithm that a kind of intra prediction block size suitable for HEVC standard divides | |
| EP3706421A1 (en) | Method and apparatus for video encoding and decoding based on affine motion compensation | |
| CN105812797A (en) | Coding unit selection method and device | |
| CN108989799A (en) | A kind of selection method, device and the electronic equipment of coding unit reference frame | |
| CN110351552B (en) | Fast coding method in video coding | |
| EP3891981A2 (en) | Method and device for picture encoding and decoding | |
| CN104581152A (en) | HEVC intra-frame prediction mode decision accelerating method | |
| CN104902276A (en) | Conversion unit diversion method and device | |
| CN102883164B (en) | A kind of decoding method of enhancement layer block unit, corresponding device | |
| CN104104947B (en) | A kind of method for video coding and device | |
| CN110149512A (en) | Inter-prediction accelerated method, control device, electronic device, computer storage medium and equipment | |
| CN105847794A (en) | HEVC intra-frame prediction mode rapid selection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |