CN104333756A - HEVC (High Efficiency Video Coding) prediction mode fast selection method based on time domain correlation - Google Patents
HEVC (High Efficiency Video Coding) prediction mode fast selection method based on time domain correlation Download PDFInfo
- Publication number
- CN104333756A CN104333756A CN201410662687.5A CN201410662687A CN104333756A CN 104333756 A CN104333756 A CN 104333756A CN 201410662687 A CN201410662687 A CN 201410662687A CN 104333756 A CN104333756 A CN 104333756A
- Authority
- CN
- China
- Prior art keywords
- prediction mode
- time domain
- optimum prediction
- unit
- coding unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses an HEVC (High Efficiency Video Coding) prediction mode fast selection method based on time domain correlation and mainly solves the problem of high computation complexity of prediction mode selection during HEVC inter-frame coding. The HEVC prediction mode fast selection method includes the following steps: (1) classifying the movement intensity of a video sequence into three states: slow movement, moderate movement and fast movement; (2) counting the probability relations of the optimum prediction modes of a coding unit and the optimum prediction modes of a coding unit adjacent to the time domain of the coding unit at the movement intensity of slow movement and moderate movement respectively; (3) building a candidate prediction mode list according to the probability relations; (4) coding inter-frame coding units according to the candidate prediction mode list to obtain the optimum prediction mode. The HEVC prediction mode fast selection method effectively reduces the computation complexity of prediction mode selection on the premise of hardly changing the video frequency distortion performance, improves the prediction mode selection speed, reduces coding time, and can be applied to real-time video compression.
Description
Technical field
The invention belongs to field of video processing, the particularly generation of interframe encode unit candidate modes in high-performance video coding HEVC, can be used for HEVC interframe encode process.
Background technology
At present, all high-quality multimedia videos application substantially, as digital video, video request program and DVD etc., the Advanced Video Coding standard also formulated for 2003 in use H.264/AVC.But along with the development that Video Applications is advanced by leaps and bounds, various intelligent terminal, as smart mobile phone, panel computer etc. have become the instrument of public recreation, for the demand of ultra high-definition video also in continuous growth, people are more and more higher for high-quality, high-resolution video requirement.In order to meet the ever-increasing demand of Video Applications, video encoding standard of new generation is arisen at the historic moment.Video coding associating group JCT-VC started in 2010 to formulate efficient video coding standard HEVC of new generation, and had formally formulated in February, 2013.When HEVC is intended to ensure that Subjective video quality is constant, compression efficiency doubles, and greatly reduces transmission video signal bandwidth.
HEVC coding framework still continues to use the hybrid encoding frame of H.26x series standard, uses classical block-based hybrid coding model, Union Movement compensation prediction, transition coding and high efficiency entropy code.Compared with video encoding standard before, HEVC coding unit adopts quad-tree structure flexibly, and large-size images block can be encoded, predicts and be converted to size, from 64 × 64 to 8 × 8, expeditiously.The experimental configuration condition of HEVC standard comprises three kinds: full I frame configuration, low time delay configuration and Stochastic accessing configuration.During Stochastic accessing configuration, adopt classification B frame, have 4 time domain layerings.
Predictive mode selection is the key technology of HEVC, its objective is and from multiple candidate modes, selects optimum prediction mode to each interframe encode unit, to obtain optimum precision of prediction.HEVC adopts rate-distortion optimization model to carry out the selection of optimum prediction mode.
In HEVC standard, the predicting mode selecting method of interframe encode unit will first to SKIP, Inter 2N × 2N, Inter 2N × N, Inter N × 2N, Inter 2N × nU, Inter 2N × nD, Inter nL × 2N, Inter nR × 2N, Intra 2N × 2N, these 10 kinds of candidate modes of Intra N × N travel through, then using the optimum prediction mode of predictive mode minimum for rate distortion costs as current coded unit.The process that this optimum prediction mode is selected makes HEVC encoder complexity sharply rise, and brings extreme difficulties to the real-time application of HEVC.Therefore need ensureing, under the prerequisite that coding efficiency is substantially constant, to reduce HEVC encoder complexity.
So far, the HEVC predictive mode fast selecting method proposed has following several:
Motion JCTVC-F045 " Early termination of CU encoding to reduce HEVC complexity ", carries out simplification judgement by the value of coded block flag CBF to predictive mode.When the method proposes the coded block flag CBF=0 of other inter-frame forecast modes except Inter N × N, the selection course of remaining predictive mode will be terminated, and no longer calculate, thus reduce encoder complexity.The method is called coded block flag quick mode CFM method.
Motion JCTVC-G543 " Early SKIP Detection for HEVC " proposes, first Searching I nter 2N × 2N pattern before detecting SKIP pattern, then motion vector difference DMV and the coded block flag CBF of Inter 2N × 2N pattern is detected, if DMV equals (0,0), and CBF equals 0, then the optimum prediction mode of current coded unit is just set as SKIP pattern, no longer remaining predictive mode is traveled through, thus greatly reduce encoder complexity.The method is called that early stage SKIP detects ESD method.
The people such as Jong-Hyeok Lee propose a kind of fast method based on time-space domain and coding depth correlation in the paper " Novel Fast PU Decision Algorithm for the HEVC Video Standard ".The method utilizes motion complexity that coding unit is divided into different motion region: if current coded unit motion complexity is less than threshold value Th1, then judge that current coded unit is slow moving region, only travels through SKIP and Inter 2N × 2N pattern; If current coded unit motion complexity is greater than threshold value Th1 and is less than threshold value Th2, then judge that current coded unit is moderate moving region, SKIP, Inter 2N × 2N, Inter 2N × N and Inter N × 2N pattern are traveled through, otherwise judge that current coded unit is fast moving region, 10 kinds of predictive modes are all traveled through, so just skip the predictive mode of redundancy, effectively reduce encoder complexity.
Above method reduces the encoder complexity of HEVC all to a certain extent.But these methods all do not carry out detailed analysis to relativity of time domain at present, HEVC predictive mode is made to select speed to fail to be promoted further.
Summary of the invention
The object of the invention is to the deficiency for above-mentioned prior art, a kind of HEVC predictive mode fast selecting method based on relativity of time domain is proposed, with when ensureing that video compression performance is substantially constant, reducing encoder complexity further, promoting predictive mode and selecting speed.
Realizing the object of the invention technological thought is: in predictive mode selection course, according to the optimum prediction mode of time domain adjacent encoder unit, provide the candidate modes of current coded unit, skip redundant prediction pattern as much as possible, then from candidate modes, optimum prediction mode is selected, improve coding rate, implementation step comprises as follows:
(1) exercise intensity of video sequence is divided into: motion is slow, moderate, motion these three kinds of states fast of moving;
(2) correlation of the optimum prediction mode of coding unit and the optimum prediction mode of its time domain adjacent encoder unit is determined:
2.1) input video sequence and experimental configuration condition, from the 2nd non-I frame, to each interframe encode unit, uses the numbering of this coding unit in reference frame, search for the coding unit with its same position, obtains time domain adjacent encoder unit;
2.2) classification of judgment experiment configuration condition:
If experimental configuration condition is low time delay, then to [20,26], [27,31], [32,36], [37,41] these four kinds different quantization parameter scopes, respectively statistics motion slowly, the optimum prediction mode of coding unit of these the two kinds of exercise intensities moderate that move and the probabilistic relation of the optimum prediction mode of its time domain adjacent encoder unit;
If experimental configuration condition is Stochastic accessing, then in 1,2,3,4 these four time domain layerings, respectively to [20,26], [27,31], [32,36], the different quantization parameter scope in [37,41] these four kinds, statistics motion slowly, the optimum prediction mode of coding unit of these the two kinds of exercise intensities moderate that move and the probabilistic relation of the optimum prediction mode of its time domain adjacent encoder unit;
(3) the experimentally optimum prediction mode of coding unit and the probabilistic relation of its time domain adjacent encoder unit optimum prediction mode in configuration condition, quantization parameter scope, exercise intensity and these situations of time domain layering, builds low time delay configuration candidate modes table and Stochastic accessing configuration candidate modes table:
3.1) optimum prediction mode selecting time domain adjacent encoder unit is respectively SKIP, Inter 2N × 2N, Inter 2N × N, Inter N × 2N, Inter 2N × nU, Inter 2N × nD, Inter nL × 2N, Inter nR × 2N, Intra 2N × 2N, each in these 10 kinds of predictive modes of Intra N × N, the optimum prediction mode calculating current coded unit is the probability of these 10 kinds of predictive modes, and sorts from big to small by probability;
3.2) from the predictive mode of maximum probability, select probability sum is not less than 90% and the predictive mode of minimum number alternatively predictive mode, and sets candidate modes sum and be no more than 5 kinds, thus builds candidate modes table;
(4) according to candidate modes table, choose candidate modes, and interframe encode unit encoded, obtain optimum prediction mode:
4.1) input video sequence and experimental configuration condition, adopts the coding unit of predicting mode selecting method to I frame and the 1st non-I frame of HEVC standard to encode, obtains the optimum prediction mode of each coding unit;
4.2) from the 2nd non-I frame, to each interframe encode unit, the classification of its exercise intensity is judged:
If the exercise intensity of current coded unit, for moving slowly, performs step 4.3); If the exercise intensity of current coded unit is moderate for moving, perform step 4.4); Otherwise, adopt the predicting mode selecting method of HEVC standard to encode to coding unit, obtain optimum prediction mode;
4.3) judge to move experimental configuration condition classification slowly:
If experimental configuration condition is low time delay, then according to the optimum prediction mode of quantization parameter QP and time domain adjacent encoder unit, low time delay configuration under move slow coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then according to the optimum prediction mode of time domain layering, quantization parameter QP and time domain adjacent encoder unit, Stochastic accessing configuration under move slow coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode;
4.4) the experimental configuration condition classification that motion is moderate is judged:
If experimental configuration condition is low time delay, then according to the optimum prediction mode of quantization parameter QP and time domain adjacent encoder unit, low time delay configuration under move moderate coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then according to the optimum prediction mode of time domain layering, quantization parameter QP and time domain adjacent encoder unit, Stochastic accessing configuration under move moderate coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode.
The present invention compared with the conventional method tool has the following advantages:
A () the present invention, due to according to relativity of time domain, has skipped the predictive mode of redundancy, make predictive mode seletion calculation complexity less, decrease a large amount of scramble times;
B () the present invention, due to when selecting candidate modes, selecting according to different motion intensity, making model selection result more accurate.
Accompanying drawing explanation
Fig. 1 is realization flow figure of the present invention;
Fig. 2 is surrounding's coding unit collection location schematic diagram of current coded unit.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.The present embodiment is implemented premised on technical solution of the present invention, gives detailed execution mode and specific operation process, but protection scope of the present invention is not limited to following embodiment.
With reference to Fig. 1, performing step of the present invention is as follows:
Step one: the exercise intensity of video sequence is classified.
1a) to current coded unit CU
0set up motion vector set { mv
1, mv
2, mv
3, mv
4, mv
5,
Wherein mv
icurrent coded unit CU
0coding unit CU around
imotion vector, mv
i=(x
i, y
i), i=1 ..., 5, x
iand y
ibe respectively abscissa and the ordinate of motion vector; CU
0coding unit set around comprises its left side adjacent encoder unit CU
1, top adjacent encoder unit CU
2, top right-hand side adjacent encoder unit CU
3, limit, upper left adjacent encoder unit CU
4with time domain adjacent encoder unit CU
5, as shown in Figure 2;
1b) the length l of each motion vector in calculating kinematical vector set:
l(mv
i)＝|x
i|+|y
i|；
1c) in step 1b) select the maximum L of length in the motion vector length set that obtains:
L＝max{l(mv
1),l(mv
2),l(mv
3),l(mv
4),l(mv
5)}，
The slow threshold value L0 of motion and motion moderate threshold value L1 1d) is set:
1d1) input video sequence and experimental configuration condition that resolution is 416 × 240;
1d2) from first non-I frame, the classification of judgment experiment configuration condition:
If experimental configuration condition is low time delay, then the motion vector of output video sequence, according to size and the video resolution of motion vector, and with reference to the threshold value T0 of motion slow state in Zeng method and the threshold value T1 of the moderate state of motion, setting threshold L0 and L1;
If experimental configuration condition is Stochastic accessing, the then motion vector of output video sequence different time domain layering, according to size and the video resolution of motion vector, and with reference to the threshold value T0 of motion slow state in Zeng method and the threshold value T1 of the moderate state of motion, the threshold value L0 of setting different time domain layering and L1;
Described Zeng method is shown in Huanqiang Zeng, the article " A Novel Fast Mode Decision for the H.264/AVC Based on Local MacroblockMotion Activity " of the people such as Canhui Cai;
1d3) repeat step 1d1) and step 1d2) obtain respectively resolution be the video sequence of 832 × 480,1280 × 720,1920 × 1080,2560 × 1600 the slow threshold value L0 of motion and motion moderate threshold value L1, obtain final different resolution video sequence to move slow threshold value L0 and the moderate threshold value L1 of motion, as table 1 and table 2:
Table 1 low time delay configuration motion slow state threshold value L0 and the moderate state threshold L1 of motion
| Video sequence resolution | Threshold value L0 | Threshold value L1 |
| 416×240 | 2 | 8 |
| 832×480、1280×720 | 3 | 12 |
| 1920×1080 | 8 | 32 |
Table 2 Stochastic accessing configuration motion slow state threshold value L0 and the moderate state threshold L1 of motion
1e) experimentally configuration condition, video resolution and time domain layering, selects the slow threshold value L0 of motion and the moderate threshold value L1 of motion, by step 1c from table 1 or table 2) in the length maximum L that obtains, compare with threshold value L0 and threshold value L1:
If L < is L0, then define exercise intensity for moving slowly;
If L0≤L < is L1, then define exercise intensity moderate for moving;
If L >=L1, then define exercise intensity fast for moving.
Step 2: the correlation determining the optimum prediction mode of coding unit and the optimum prediction mode of its time domain adjacent encoder unit.
2a) input video sequence BQSquare and experimental configuration condition, from the 2nd non-I frame, to each interframe encode unit, uses the numbering of this coding unit in reference frame, search for the coding unit with its same position, obtains time domain adjacent encoder unit;
2b) the classification of judgment experiment configuration condition:
If experimental configuration condition is low time delay, then front 100 frames of test video sequence BQSquare, to [20,26], [27,31], [32,36], the different quantization parameter scope in [37,41] these four kinds, respectively statistics motion slowly, the coding unit optimum prediction mode of these the two kinds of exercise intensities moderate that move and the probabilistic relation of its time domain adjacent encoder unit optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then front 65 frames of test video sequence BQSquare, in 1,2,3,4 these four time domain layerings, respectively to [20,26], [27,31], [32,36], [37,41] these four kinds different quantization parameter scopes, statistics motion slowly, the optimum prediction mode of coding unit of these the two kinds of exercise intensities moderate that move and the probabilistic relation of the optimum prediction mode of its time domain adjacent encoder unit;
2c) repeat step 2a) and step 2b) obtain the probabilistic relation of the optimum prediction mode of the coding unit of video sequence BasketballPass and Johnny and the optimum prediction mode of its time domain adjacent encoder unit respectively, do average to the probabilistic relation of these three video sequences above-mentioned, obtain the probabilistic relation of final low time delay configuration and the lower coding unit of different motion intensity of Stochastic accessing configuration and the optimum prediction mode of its time domain adjacent encoder unit, only providing the lower quantization parameter scope of low time delay configuration is below [20, 26] probabilistic relation time, and the lower time domain of Stochastic accessing configuration is layered as 1 and quantization parameter scope is [20, 26] probabilistic relation, if table 3 is to table 6:
The probabilistic relation of the optimum prediction mode of the table 3 low time delay configuration slow coding unit of motion and time domain adjacent encoder unit
The probabilistic relation of the optimum prediction mode of the table 4 low time delay configuration moderate coding unit of motion and time domain adjacent encoder unit
The probabilistic relation of the optimum prediction mode of the table 5 Stochastic accessing configuration slow coding unit of motion and time domain adjacent encoder unit
The probabilistic relation of the optimum prediction mode of the table 6 Stochastic accessing configuration moderate coding unit of motion and time domain adjacent encoder unit
In above-mentioned table 3-table 6, pattern 0 represents SKIP, pattern 1 represents Inter 2N × 2N, and pattern 2 represents Inter 2N × N, and mode 3 represents Inter N × 2N, pattern 4 represents Inter 2N × nU, pattern 5 represents Inter 2N × nD, and pattern 6 represents Inter nL × 2N, and mode 7 represents Inter nR × 2N, pattern 8 represents Intra 2N × 2N, and pattern 9 represents Intra N × N.
Step 3: according to table 3-table 6, builds low time delay configuration candidate modes table and Stochastic accessing configuration candidate modes table.
3a) optimum prediction mode of coding unit and its time domain adjacent encoder unit comprises: SKIP, Inter 2N × 2N, Inter 2N × N, Inter N × 2N, Inter 2N × nU, Inter 2N × nD, Inter nL × 2N, these 10 kinds of predictive modes of Inter nR × 2N, Intra 2N × 2N, Intra N × N; The optimum prediction mode selecting time domain adjacent encoder unit is SKIP pattern, and the optimum prediction mode calculating current coded unit is respectively the probability of above-mentioned 10 kinds of predictive modes, and sorts from big to small by probability;
3b) repeating step 3a) optimum prediction mode of selecting time domain adjacent encoder unit respectively be other 9 kinds of predictive modes, the optimum prediction mode of calculating current coded unit is the probability of above-mentioned 10 kinds of predictive modes, and sorts from big to small by probability;
3c) from the predictive mode of maximum probability, select probability sum is not less than 90% and the predictive mode of minimum number alternatively predictive mode, and sets candidate modes sum and be no more than 5 kinds, thus builds candidate modes table, as table 7-table 10;
The slow coding unit candidate modes table of table 7 low time delay configuration motion
The moderate coding unit candidate modes table of table 8 low time delay configuration motion
The slow coding unit candidate modes table of table 9 Stochastic accessing configuration motion
The moderate coding unit candidate modes table of table 10 Stochastic accessing configuration motion
Step 4: according to table 7-table 10, choose candidate modes, in his-and-hers watches 11, the interframe encode unit of each video sequence is encoded, and obtains optimum prediction mode.
4a) input video sequence, adopts the coding unit of predicting mode selecting method to I frame and the 1st non-I frame of HEVC standard to encode, obtains the optimum prediction mode of each coding unit;
4b) from the 2nd non-I frame, to each interframe encode unit, judge its exercise intensity:
If the exercise intensity of current coded unit, for moving slowly, performs step 4c); If the exercise intensity of current coded unit is moderate for moving, perform step 4d); Otherwise, adopt the predicting mode selecting method of HEVC standard to encode to coding unit, obtain optimum prediction mode;
4c) judge to move experimental configuration condition classification slowly:
If experimental configuration condition is low time delay, then according to the optimum prediction mode of quantization parameter QP and time domain adjacent encoder unit, selects corresponding candidate modes to travel through in table 7, obtain optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then according to the optimum prediction mode of time domain layering, quantization parameter QP and time domain adjacent encoder unit, in table 9, selects corresponding candidate modes to travel through, obtain optimum prediction mode;
4d) judge the experimental configuration condition classification that motion is moderate:
If experimental configuration condition is low time delay, then according to the optimum prediction mode of quantization parameter QP and time domain adjacent encoder unit, selects corresponding candidate modes to travel through in table 8, obtain optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then according to the optimum prediction mode of time domain layering, quantization parameter QP and time domain adjacent encoder unit, selects corresponding candidate modes to travel through in table 10, obtain optimum prediction mode.
Effect of the present invention further illustrates by following emulation:
1. experimental situation
Use VS2010 coding environment, test with reference software HM16.0, experimental configuration condition is low time delay configuration and Stochastic accessing configuration.
The video sequence details of experiment test are as table 11:
Table 11 video sequence details
2. experiment content
The inventive method, CFM fast method, ESD fast method and the Lee fast method all video sequences respectively in his-and-hers watches 11 are used to encode, record coding time and distortion performance estimator BD-PSNR.The present invention compares with the coding efficiency of these three kinds of fast methods as shown 12-table 14, wherein table 12 is the inventive method and the comparing of CFM fast method coding efficiency, table 13 is the inventive method and the comparing of ESD fast method coding efficiency, and table 14 is the inventive method and the comparing of Lee fast method coding efficiency.
Wherein, show in 12-table 14
represent the inventive method time variation amount compared with existing fast selecting method, "-" represents that the inventive method has raised speed in the time than existing fast method.BD-PSNR represents under given equal code check, the difference of the brightness peak signal to noise ratio PSNR-Y of two kinds of methods, and its unit is dB, and "-" represents that the inventive method reduces than existing fast method PSNR-Y.
Table 12 the inventive method compares with CFM fast method
From table 12, the inventive method is compared with CFM fast method, during low time delay configuration when BD-PSNR on average reduces 0.07dB, scramble time has on average raised speed 4.69%, during Stochastic accessing configuration when BD-PSNR on average reduces 0.07dB, the scramble time has on average raised speed 8.43%.
Table 13 the inventive method compares with ESD fast method
From table 13, the inventive method is compared with ESD fast method, during low time delay configuration when BD-PSNR on average reduces 0.095dB, scramble time has on average raised speed 13.81%, during Stochastic accessing configuration when BD-PSNR on average reduces 0.10dB, the scramble time has on average raised speed 11.68%.
Table 14 the inventive method compares with Lee fast method
From table 14, the inventive method is compared with Lee fast method, and during Stochastic accessing configuration when BD-PSNR on average reduces 0.04dB, the scramble time has on average raised speed 4.22%.
To sum up, the present invention utilizes relativity of time domain, skips the predictive mode of redundancy, when BD-PSNR is substantially identical, improves the speed that predictive mode is selected further.
Foregoing description is preferred embodiment of the present invention, and obvious researcher in this field can make various amendment and replacement with reference to preferred embodiment of the present invention and accompanying drawing to the present invention, and these amendments and replacement all should fall within protection scope of the present invention.
Claims (2)
1., based on a HEVC predictive mode fast selecting method for relativity of time domain, it is characterized in that, comprise the steps:
(1) exercise intensity of video sequence is divided into: motion is slow, moderate, motion these three kinds of states fast of moving;
(2) correlation of the optimum prediction mode of coding unit and the optimum prediction mode of its time domain adjacent encoder unit is determined:
2.1) input video sequence and experimental configuration condition, from the 2nd non-I frame, to each interframe encode unit, uses the numbering of this coding unit in reference frame, search for the coding unit with its same position, obtains time domain adjacent encoder unit;
2.2) classification of judgment experiment configuration condition:
If experimental configuration condition is low time delay, then to [20,26], [27,31], [32,36], [37,41] these four kinds different quantization parameter scopes, respectively statistics motion slowly, the optimum prediction mode of coding unit of these the two kinds of exercise intensities moderate that move and the probabilistic relation of the optimum prediction mode of its time domain adjacent encoder unit;
If experimental configuration condition is Stochastic accessing, then in 1,2,3,4 these four time domain layerings, respectively to [20,26], [27,31], [32,36], the different quantization parameter scope in [37,41] these four kinds, statistics motion slowly, the optimum prediction mode of coding unit of these the two kinds of exercise intensities moderate that move and the probabilistic relation of the optimum prediction mode of its time domain adjacent encoder unit;
(3) the experimentally optimum prediction mode of coding unit and the probabilistic relation of its time domain adjacent encoder unit optimum prediction mode in configuration condition, quantization parameter scope, exercise intensity and these situations of time domain layering, builds low time delay configuration candidate modes table and Stochastic accessing configuration candidate modes table:
3.1) optimum prediction mode selecting time domain adjacent encoder unit is respectively SKIP, Inter 2N × 2N, Inter2N × N, Inter N × 2N, Inter 2N × nU, Inter 2N × nD, Inter nL × 2N, Inter nR × 2N, Intra2N × 2N, each in these 10 kinds of predictive modes of Intra N × N, the optimum prediction mode calculating current coded unit is the probability of these 10 kinds of predictive modes, and sorts from big to small by probability;
3.2) from the predictive mode of maximum probability, select probability sum is not less than 90% and the predictive mode of minimum number alternatively predictive mode, and sets candidate modes sum and be no more than 5 kinds, thus builds candidate modes table;
(4) according to candidate modes table, choose candidate modes, and interframe encode unit encoded, obtain optimum prediction mode:
4.1) input video sequence and experimental configuration condition, adopts the coding unit of predicting mode selecting method to I frame and the 1st non-I frame of HEVC standard to encode, obtains the optimum prediction mode of each coding unit;
4.2) from the 2nd non-I frame, to each interframe encode unit, the classification of its exercise intensity is judged:
If the exercise intensity of current coded unit, for moving slowly, performs step 4.3); If the exercise intensity of current coded unit is moderate for moving, perform step 4.4); Otherwise, adopt the predicting mode selecting method of HEVC standard to encode to coding unit, obtain optimum prediction mode;
4.3) judge to move experimental configuration condition classification slowly:
If experimental configuration condition is low time delay, then according to the optimum prediction mode of quantization parameter QP and time domain adjacent encoder unit, low time delay configuration under move slow coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then according to the optimum prediction mode of time domain layering, quantization parameter QP and time domain adjacent encoder unit, Stochastic accessing configuration under move slow coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode;
4.4) the experimental configuration condition classification that motion is moderate is judged:
If experimental configuration condition is low time delay, then according to the optimum prediction mode of quantization parameter QP and time domain adjacent encoder unit, low time delay configuration under move moderate coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode;
If experimental configuration condition is Stochastic accessing, then according to the optimum prediction mode of time domain layering, quantization parameter QP and time domain adjacent encoder unit, Stochastic accessing configuration under move moderate coding unit candidate modes table in select corresponding candidate modes to travel through, obtain optimum prediction mode.
2. as claimed in claim 1 based on the HEVC predictive mode fast selecting method of relativity of time domain, it is characterized in that: the exercise intensity of video sequence is divided in (1) by described step: motion is slow, moderate, motion these three kinds of states fast of moving, and carries out as follows:
1a) to current coded unit CU
0set up motion vector set { mv
1, mv
2, mv
3, mv
4, mv
5,
Wherein mv
icurrent coded unit CU
0coding unit CU around
imotion vector, mv
i=(x
i, y
i), i=1 ..., 5, x
iand y
ibe respectively abscissa and the ordinate of motion vector; CU
0coding unit around comprises its left side adjacent encoder unit CU
1, top adjacent encoder unit CU
2, top right-hand side adjacent encoder unit CU
3, limit, upper left adjacent encoder unit CU
4with time domain adjacent encoder unit CU
5;
1b) the length l of each motion vector in calculating kinematical vector set:
l(mv
i)＝|x
i|+|y
i|，
1c) in step 1b) select the maximum L of length in the motion vector length set that obtains:
L＝max{l(mv
1),l(mv
2),l(mv
3),l(mv
4),l(mv
5)}，
1d) by step 1c) the length maximum L that obtains compares with the slow threshold value L0 of motion and the moderate threshold value L1 that moves:
If L < is L0, then define exercise intensity for moving slowly;
If L0≤L < is L1, then define exercise intensity moderate for moving;
If L >=L1, then define exercise intensity motion fast.
Wherein, move slow threshold value L0 and the moderate threshold value L1 of motion, according to size and the video resolution of motion vector, and determines with reference to Zeng method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410662687.5A CN104333756B (en) | 2014-11-19 | 2014-11-19 | HEVC predictive mode fast selecting methods based on relativity of time domain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410662687.5A CN104333756B (en) | 2014-11-19 | 2014-11-19 | HEVC predictive mode fast selecting methods based on relativity of time domain |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104333756A true CN104333756A (en) | 2015-02-04 |
| CN104333756B CN104333756B (en) | 2017-10-24 |
Family
ID=52408395
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410662687.5A Active CN104333756B (en) | 2014-11-19 | 2014-11-19 | HEVC predictive mode fast selecting methods based on relativity of time domain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104333756B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104902271A (en) * | 2015-05-15 | 2015-09-09 | 腾讯科技（北京）有限公司 | Prediction mode selection method and device |
| CN106170089A (en) * | 2016-08-25 | 2016-11-30 | 上海交通大学 | Based on multiplex coding system and method H.265 |
| CN106331699A (en) * | 2015-06-19 | 2017-01-11 | 华为技术有限公司 | Image coding method and image encoder |
| CN107566846A (en) * | 2017-09-20 | 2018-01-09 | 北京维境视讯信息技术有限公司 | Video coding skip mode decision-making technique, device, equipment and storage medium |
| CN108989801A (en) * | 2018-07-27 | 2018-12-11 | 福州大学至诚学院 | A kind of adaptive model selection fast method based on space-time characterisation |
| CN112839224A (en) * | 2019-11-22 | 2021-05-25 | 腾讯科技（深圳）有限公司 | Prediction mode selection method and device, video coding equipment and storage medium |
| CN113596456A (en) * | 2019-09-23 | 2021-11-02 | 杭州海康威视数字技术股份有限公司 | Encoding and decoding method, device and equipment |
| US11363298B2 (en) | 2020-08-03 | 2022-06-14 | Wiston Corporation | Video processing apparatus and processing method of video stream |
| US11880966B2 (en) | 2020-08-03 | 2024-01-23 | Wistron Corporation | Image quality assessment apparatus and image quality assessment method thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102984521A (en) * | 2012-12-12 | 2013-03-20 | 四川大学 | High-efficiency video coding inter-frame mode judging method based on temporal relativity |
| US20130336384A1 (en) * | 2012-06-15 | 2013-12-19 | General Instrument Corporation | Method and apparatus for efficient slice header processing |
-
2014
- 2014-11-19 CN CN201410662687.5A patent/CN104333756B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130336384A1 (en) * | 2012-06-15 | 2013-12-19 | General Instrument Corporation | Method and apparatus for efficient slice header processing |
| CN102984521A (en) * | 2012-12-12 | 2013-03-20 | 四川大学 | High-efficiency video coding inter-frame mode judging method based on temporal relativity |
Non-Patent Citations (3)
| Title |
|---|
| LEE J H等: "Novel fast PU decision algorithm for the HEVC video standard", 《IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING》 * |
| ZENG HUANQIANG等: "A Novel Fast Mode Decision for the H.264/AVC Based on Local Macroblock Motion Activity", 《IEEE INTERNATIONAL CONFERENCE ON IMAGE AND GRAPHICS》 * |
| 许东旭等: "高效率视频编码帧内快速算法", 《计算机应用》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104902271B (en) * | 2015-05-15 | 2017-12-29 | 腾讯科技（北京）有限公司 | Predicting mode selecting method and device |
| CN104902271A (en) * | 2015-05-15 | 2015-09-09 | 腾讯科技（北京）有限公司 | Prediction mode selection method and device |
| CN106331699B (en) * | 2015-06-19 | 2019-04-26 | 华为技术有限公司 | Image encoding method and image encoder |
| CN106331699A (en) * | 2015-06-19 | 2017-01-11 | 华为技术有限公司 | Image coding method and image encoder |
| CN106170089A (en) * | 2016-08-25 | 2016-11-30 | 上海交通大学 | Based on multiplex coding system and method H.265 |
| CN106170089B (en) * | 2016-08-25 | 2020-05-22 | 上海交通大学 | H.265-based multi-path coding method |
| CN107566846A (en) * | 2017-09-20 | 2018-01-09 | 北京维境视讯信息技术有限公司 | Video coding skip mode decision-making technique, device, equipment and storage medium |
| CN107566846B (en) * | 2017-09-20 | 2020-04-14 | 北京字节跳动网络技术有限公司 | Video coding skipping mode decision method, device, equipment and storage medium |
| CN108989801A (en) * | 2018-07-27 | 2018-12-11 | 福州大学至诚学院 | A kind of adaptive model selection fast method based on space-time characterisation |
| CN108989801B (en) * | 2018-07-27 | 2021-10-08 | 福州大学至诚学院 | Space-time characteristic-based adaptive mode selection rapid method |
| CN113596456A (en) * | 2019-09-23 | 2021-11-02 | 杭州海康威视数字技术股份有限公司 | Encoding and decoding method, device and equipment |
| CN112839224A (en) * | 2019-11-22 | 2021-05-25 | 腾讯科技（深圳）有限公司 | Prediction mode selection method and device, video coding equipment and storage medium |
| US11363298B2 (en) | 2020-08-03 | 2022-06-14 | Wiston Corporation | Video processing apparatus and processing method of video stream |
| US11880966B2 (en) | 2020-08-03 | 2024-01-23 | Wistron Corporation | Image quality assessment apparatus and image quality assessment method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104333756B (en) | 2017-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104333756A (en) | HEVC (High Efficiency Video Coding) prediction mode fast selection method based on time domain correlation | |
| CN102984521B (en) | High-efficiency video coding inter-frame mode judging method based on temporal relativity | |
| CN103873861B (en) | Coding mode selection method for HEVC (high efficiency video coding) | |
| CN105141954B (en) | A kind of HEVC interframe encodes fast schema selection method | |
| CN104754357B (en) | Intraframe coding optimization method and device based on convolutional neural networks | |
| CN100574447C (en) | Fast intraframe predicting mode selecting method based on the AVS video coding | |
| CN104394409B (en) | HEVC predictive mode fast selecting methods based on spatial correlation | |
| CN105306947B (en) | video transcoding method based on machine learning | |
| CN104539962A (en) | Layered video coding method fused with visual perception features | |
| CN105791826A (en) | Data mining-based HEVC inter-frame fast mode selection method | |
| CN103997645B (en) | Quick HEVC intra-frame coding unit and pattern decision method | |
| CN102763411A (en) | Method and apparatus for encoding video, and method and apparatus for decoding video | |
| CN103384325A (en) | Quick inter-frame prediction mode selection method for AVS-M video coding | |
| CN104883565A (en) | Decision-making method and device for intra-frame prediction mode of high efficiency video coding | |
| CN105657420A (en) | HEVC-oriented fast intra-frame prediction mode decision method and device | |
| CN104853191A (en) | HEVC fast coding method | |
| CN104168480A (en) | Fast selecting method for intra-frame prediction encoding modes based on HEVC standard | |
| CN101304529A (en) | Method and device for selecting macro block pattern | |
| CN106937116A (en) | Low-complexity video coding method based on random training set adaptive learning | |
| CN103533355A (en) | Quick coding method for HEVC (high efficiency video coding) | |
| CN103888763B (en) | Intra-frame coding method based on HEVC | |
| CN104333754A (en) | SHVC (scalable high efficiency video coding) enhancement layer video coding method based on rapid prediction mode selection | |
| CN104581152A (en) | HEVC intra-frame prediction mode decision accelerating method | |
| CN101895761B (en) | Quick intraframe prediction algorithm | |
| CN103702131B (en) | Pattern-preprocessing-based intraframe coding optimization method and system |
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 |