CN103384325A - Quick inter-frame prediction mode selection method for AVS-M video coding - Google Patents
Quick inter-frame prediction mode selection method for AVS-M video coding Download PDFInfo
- Publication number
- CN103384325A CN103384325A CN2013100569696A CN201310056969A CN103384325A CN 103384325 A CN103384325 A CN 103384325A CN 2013100569696 A CN2013100569696 A CN 2013100569696A CN 201310056969 A CN201310056969 A CN 201310056969A CN 103384325 A CN103384325 A CN 103384325A
- Authority
- CN
- China
- Prior art keywords
- mode
- skip
- search
- avs
- pattern
- 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
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention provides a quick inter-frame prediction mode selection method for AVS-M video coding. Algorithms can be classified according to the distribution situation of inter-frame coding modes and are arranged according to processing priority, different types are judged sequentially, possible modes are compared selectively, and the number of modes to be judged is reduced. Early termination judgment is utilized to carry out priority judgment on an SKIP mode, and whether other modes are early termination is judged through threshold value judgment. If threshold value judgment conditions are met, the SKIP mode is directly utilized to code. If the threshold value judgment conditions are not met, an Inter4*4 mode, an Inter 8*8 mode and an Inter 16*16 mode are used as initial modes. A part of macro block modes are removed according to whether rate-distortion cost meets monotony, and accordingly useless block mode search processes are reduced. Compared with a full-mode selection algorithm, the calculating times is reduced by 52.5% to 85.8%, and the calculating complexity of inter-frame forecasting modules is reduced.
Description
Technical field
The present invention relates to the technical field of video coding in the signal processing, be specifically related to a kind of fast intraframe predicting mode selecting method of AVS-M Video coding.
Background technology
The digital audio/video industry is one of information industry three large parts, the digital audio/video industry size of China is huge, but intellectual property is weak, and this field adopts these standards all to need to pay high patent fee for a long time by MPEG-2, MPEG-4 with H.264 wait foreign technology standard monopolization.Be the patent fee burden that reduces domestic audio frequency and video industry, the core competitiveness that promotes domestic enterprise, the Ministry of Information industry is in June, 2002 approval establishment " digital audio/video coding techniques standard operation group ", be engaged in scientific research institution and the enterprise of the research and development of digital audio/video coding techniques in the United Nations, demand for domestic audio frequency and video industry, proposed the autonomous information source coding standard of China-" information technology advanced audio/video coding " series standard, be called for short AVS(Audio Video coding Standard).AVS-M is AVS1-Part7, is that China is towards the video encoding standard of mobile communication.
AVS with H.264 compare, have 4 large advantages: (1) performance is high, code efficiency is 2.4 times of MPEG-2, with H.264 suitable; (2) complexity is low, encoder complexity be equivalent to H.264 70%, decoding complex degree be equivalent to H.264 30%; (3) software and hardware realizes that cost is all lower than H.264; (4) patent fee is low, only collects the software and hardware patent fee of 1 yuan.The use of AVS not only can annual patent fee of saving tens dollars, can also drive the chip industry development of China, accelerates to solve the problem of China's IT enterprises coreless technology.
The inter prediction technology is the key algorithm in the AVS-M standard, AVS-M employing rate-distortion optimization RDO(Rate Distortion Optimization) syntype system of selection, relatively under all patterns the coding gained rate distortion costs RDC(Rate Distortion Cost), therefrom the pattern of selection rate distortion cost minimum as the coding mode of optimum, has greatly increased encoder complexity when improving code efficiency.Adopt the inter prediction module of the whole coded block patterns of multi-reference frame to account for more than 80% of whole AVS-M encoding calculation amount, use in real time the computation complexity that must reduce this module.
Present many scholars have carried out a large amount of research to interframe prediction mode selection algorithm, and the fast mode decision algorithm can be divided into following two classes substantially: a class is based on the method that raw video signal feature (edge, texture and interframe variation etc.) is analyzed; The another kind of method of the coding mode correlation analysis of the adjacent macro block in the timely Kongxiang of coding mode gained information that is based on.Typical algorithm has: based on the algorithm of marginal information, utilize edge strength to detect the stronger macro block of Space Consistency, and the difference of utilizing current macro and former frame correspondence position macro block judges whether this macro block is static, the macro block of or time domain instantaneous static strong for Space Consistency is only selected the larger-size pattern of piece.The method need to be pursued pixel edge calculation information, and determines in algorithm that suitable threshold ratio is more difficult.At first the algorithm of down-sampled filtering adopted the mean filter method to carry out down-sampled filtering to 16 * 16 macro blocks of same position in original brightness 16 * 16 macro blocks of present encoding and former frame original brightness data before motion search, then calculated theirs
SADValue is with it
SADValue compares with the thresholding that passes through statistical analysis as the similarity reference of front and back two macro blocks, judges whether current block is static block, selects in advance the optimal prediction modes coding.Because need calculate after down-sampled filtering
SADValue has reduced code efficiency.
Summary of the invention
For the video encoding standard AVS-M needs more efficient interframe prediction mode selection algorithm of China towards mobile communication, in order to solve the deficiencies in the prior art, the present invention discloses a kind of quick interframe prediction mode selection algorithm of pattern classification of the AVS-M of being applicable to Video coding, compare with the syntype selection algorithm, this algorithm is when having kept well code efficiency and reconstructed image quality, and coding rate can be brought up to 2.10～7.04 times of syntype selection algorithm.
For achieving the above object, the present invention adopts following fast intraframe predicting mode selecting method: as shown in Figure 1, (1) according to the interframe encoding mode distribution situation, it is classified, arrange by processing priority, adjudicate dissimilar in order, relatively more possible pattern, effectively reduce the quantity that needs decision pattern selectively, determines fast optimization model.(2) adopt the premature termination judgement, the SKIP pattern is preferentially adjudicated, whether other pattern passing threshold adjudicates premature termination, if satisfy the threshold value judgment condition, judges that directly this macro block is with the SKIP pattern-coding.If do not satisfy the threshold value judgment condition, with Inter4 * 4, Inter8 * 8, Interl6 * 16 3 kind of pattern is as originate mode, whether meet monotonicity according to the rate distortion costs of these three kinds of patterns, first get rid of a part of macro block mode, reduce useless block mode search procedure, if rate distortion costs does not meet monotonicity, the block mode of search overall dimension.
Compared with the prior art, advantage of the present invention and good effect are: the present invention classifies interframe encoding mode, arrange by processing priority, adjudicate dissimilar in order, selectively relatively may pattern, and adopt the premature termination judgement, when effectively minimizing needs decision pattern quantity, can determine fast optimization model.Compare with the syntype selection algorithm, calculation times reduces 52.5%～85.8%, can greatly reduce the computation complexity of inter prediction module, improves coding rate.
The below describes in detail method for classifying modes, premature termination judgement and this three part of mode adjudging of employing of the present invention.
Method for classifying modes.
The statistical analysis of interframe encode optimal mode selection result.
The present invention has tested some different texture features and motion feature on AVS-M Knowledge Verification Model software WM3.3a sequence, table 1 listed wherein in the P frame of 3 representative sequences different quantization parameters (
QP) time each pattern distribution proportion.
As can be seen from Table 1, the final block mode of selecting is relevant with textural characteristics with the motion feature of sequence: adopt the probability of intra prediction mode less than 0.1% in the P frame; Simple for texture, the sequence A kiyo that change of background is little and News, SKIP and relatively large pattern proportion are larger; And to adopt more little block mode for the sequence of high-speed motion, texture complexity, and details is abundanter, and little block mode is more.Test also shows the increase along with quantization parameter, adopts the probability of the relatively large patterns such as Interl6 * 16, Interl6 * 8, Inter8 * 16 obviously to increase.
The different sequences of table 1 are chosen the pattern percentage
The mode profile rule shows, if only select the useful block mode of minimizing prediction residual value, the block mode search procedure of skip useless, the motion vector that obtains so both can improve picture quality, guarantee the advantage of multi-mode motion estimation algorithm, can effectively reduce algorithm complex again.
The inter-frame forecast mode classification.
The P frame encoding mode that the AVS-M video encoding standard adopts comprises 8 kinds of inter-frame modes and 2 kinds of frame modes.As shown in Figure 2, interframe encoding mode is respectively Inter16 * 16, Inter16 * 8, Inter8 * 16, Inter8 * 8, and for Inter8 * 8 patterns, each 8 * 8 sub-block can be divided into again Inter8 * 4, Inter4 * 8, Inter4 * 4 patterns.In addition, also adopt the SKIP pattern, with the large-area stagnant zone of effectively encoding zone consistent with motion.Intra-frame encoding mode is divided into Intra-Direct and Intra-4 * 4 two type according to prediction piece size.Intra-Direct is fit to image flat site coding take 16 * 16 luminance block as the prediction unit.9 kinds of predictive modes are supported take 4 * 4 luminance block as the prediction unit in Intra-4 * 4, are fit to the macro block of texture complexity.
The present invention classifies it according to AVS-M interframe encoding mode distribution situation, and SKIP pattern computing wherein simply and not needs coding, is divided into separately a class; Interl6 * 16, Interl6 * 8, Inter8 * 16 are divided into the macro block class; Inter8 * 8, Inter8 * 4, Inter4 * 8, Inter4 * 4 are divided into the sub-block class; Frame mode is as a class.These types are arranged by processing priority: { SKIP}, { Interl6 * 16, Interl6 * 8, Inter8 * 16 }, { Inter8 * 8, Inter8 * 4, Inter4 * 8, Inter4 * 4 }, { Intra-Direct, Intra-4 * 4 }.Algorithm of the present invention is adjudicated dissimilar in order, and relatively more possible pattern, make when effectively minimizing needs decision pattern quantity selectively, can determine fast optimization model.
The premature termination judgement of pattern.
The SKIP pattern need not encoded, and directly copies the reference macroblock of former frame, and characteristics are: (1) coded-bit is 0; (2) motion vector is (0,0), perhaps equals predictive vector.Therefore its process computation complexity is minimum and simple, can preferentially adjudicate it, and judges whether premature termination of other mode adjudgings by some conditions.
For static background is arranged, less video sequence moves, most of macro block can be encoded with the SKIP pattern, namely do not transmit the information about motion vector and predicted residual signal, residual error thinks and is quantified as zero, can utilize in advance prediction residual between macroblock to be encoded and reference macroblock according to formula (1)
SADJudge whether to some extent motion of macro block.
In formula,
c(
x,
y) be the pixel value of reference macroblock,
s(
x,
y) be the pixel value of current macro, threshold value
THDetermine according to full zero-detection threshold value in AVS-M:
At first calculate the prediction residual of SKIP pattern, if satisfy formula (1), the reference block under current block and SKIP pattern can mate well, no longer carries out the estimation of other pattern, judges that directly this macro block is with the SKP pattern-coding.
Mode adjudging.
The judgement of inter-frame mode.
In order to select forced coding pattern, AVS-M to adopt rate-distortion optimization technology RDO to carry out the selection of optimization model from above-mentioned numerous patterns.Model selection algorithm based on RDO passes through all interframe of traversal and intra-frame encoding mode, chooses the pattern of rate distortion costs minimum as the forced coding pattern, and the rate distortion costs function definition is as follows:
In formula
J ModeFor
RDCost RFor with institute's lectotype and macro block quantized value (
QP) relevant bit number, comprise the bit of macro block head, motion vector and all transform blocks;
λ Mode=0.85 * 2
(
QP-11)/4
Be the rate distortion parameter relevant to quantization parameter;
SSDBe the vision signal of original macro and the error sum of squares between reconstruction macro block vision signal, namely
Reconstructed blocks element value in formula
f RDO(
X, y) need to be to residual block through obtaining with the addition of prediction piece again after variation → quantification → inverse quantization → inverse transformation process.
The present invention is Inter4 * 4, Inter8 * 8, and Interl6 * 16 3 kind of pattern is as originate mode, whether meets monotonicity according to the rate distortion costs of these three kinds of patterns, first gets rid of a part of macro block mode, thereby reduces useless block mode search procedure.
If
J Mode(16 * 16)＞
J Mode(8 * 8)＞
J Mode(4 * 4) or
J Mode(16 * 16)＜
J Mode(8 * 8)＜
J Mode(4 * 4), rate distortion costs is dull.If rate distortion costs has monotonicity, only need the pattern between two kinds of optimal modes of check, for example: 8 * 8,4 * 4 is optimal mode, illustrates that macro block trends towards using small size, only need detect 8 * 4 and 4 * 8 patterns again; 16 * 16,8 * 8 is optimal mode, illustrates that macro block trends towards using large scale, only need detect 16 * 8 and 8 * 16 patterns again.If rate distortion costs does not have monotonicity, to check all patterns.
The decision rule that the present invention takes inter mode decision is as follows: at first search for 16 * 16 and 8 * 8, calculate
J Mode(16 * 16) and
J Mode(8 * 8), if
J Mode(SKIP) simultaneously less than
J Mode(16 * 16) and
J Mode(8 * 8), account for motion are estimated and are compensated the effect that rear its match condition is difficult to surpass the SKIP pattern, select the SKIP pattern as the forced coding pattern, and stop search.Otherwise search for 4 * 4, calculate
J Mode(4 * 4), relatively
J Mode(16 * 16),
J Mode(8 * 8) and
J ModeThe monotonicity of (4 * 4).If
J Mode(16 * 16) maximum continues search 8 * 4 and 4 * 8, and calculates respectively
J Mode(8 * 4) and
J Mode(4 * 8), relatively
J Mode(8 * 4),
J Mode(4 * 8) and
J Mode(4 * 4); If
J Mode(4 * 4) maximum continues search 16 * 8 and 8 * 16, and calculates respectively
J Mode(16 * 8) and
J Mode(8 * 16), relatively
J Mode(16 * 16),
J Mode(16 * 8) and
J Mode(8 * 16).If rate distortion costs does not meet monotonicity, the block mode of search overall dimension.
The judgement of frame mode.
For improving interframe encode efficient, AVS-M allows macro block to adopt frame mode when interframe encode.Experiment shows that the optimal mode that only has the only a few macro block is frame mode, if calculate frame mode one time during each macroblock coding, can increase the computational complexity of coding.The present invention is according to the time domain spatial correlation of video image, with the optimum rate distortion costs of adjacent macroblocks, one decision threshold is set
TIf,
J Mode＞
TCalculate frame mode; Otherwise do not calculate.
J Mode(inter) be minimum rate distortion costs after inter prediction,
TBe defined as follows:
As shown in Figure 3, X
0Be current coding macro block, X
1For in former frame with X
0The macro block of same position,
J Mode(A
0),
J Mode(B
0),
J Mode(C
0),
J Mode(X
1) represent respectively A
0, B
0, C
0And X
1Optimum rate distortion costs.:
Description of drawings
Fig. 1 is the technical solution used in the present invention schematic diagram.
Fig. 2 is the inter-frame forecast mode of 7 kinds of sizes of AVS-M.
Fig. 3 is the time-space domain correlation schematic diagram of image.
Fig. 4 is Fast inter predication mode decision algorithm flow chart of the present invention.
Embodiment
Fast inter predication mode decision algorithm flow process of the present invention as shown in Figure 4, the concrete Fast inter predication mode decision algorithm step of implementing is as follows.
Step 3
J Modo(SKIP)＜
J Modo(16 * 16),
J Modo(SKIP)＜
J Modo(8 * 8), SKIP is optimal mode, forwards step 7 to, otherwise, search 4 * 4.
Step 4
J Modo(16 * 16)＞
J Modo(8 * 8)＞
J ModoWhen (4 * 4), search 8 * 4 and 4 * 8 forwards step 5 to,
J Modo(16 * 16)＜
J Modo(8 * 8)＜
J ModoWhen (4 * 4), search 8 * 4 and 4 * 8,16 * 8 and 8 * 16 forwards step 5 to; Otherwise, search 16 * 8 and 8 * 16.
Step 6
J Modo(inter)＞
T, detect frame mode, determine optimal mode, otherwise, forward step 7 to.
The present invention select to move less Akiyo, medium News and each 30 frames of three typical QCIF format standard image sequences of violent Foreman that move of motion are tested algorithm of the present invention and AVS-M Knowledge Verification Model software WM3.3a and down-sampled filtering algorithm performance.Type of video sequence: IPPP.Reference frame number is 1, and the rate-distortion optimization option is set to 1, and coding rate is set to 30fps, the coding parameter of employing: the hunting zone is 16, and search precision is 1/4 pixel.Test platform: CPU is Intel Pentium4 3.0G, in save as DDR512MB, operating system is Windows XP2.
The calculation times of table 2 algorithm relatively
Table 2 has provided the algorithm of the coding mode selection algorithm of WM3.3a, down-sampled filtering and the comparison of fast selection algorithm calculation times of the present invention, can find out,
QP=40 o'clock, this paper algorithm is compared with primal algorithm, calculation times has reduced 52.5%～85.8%, this paper algorithm is compared with the algorithm of down-sampled filtering, calculation times has reduced 9.25%～22.98%, this depends on two aspects: the first, and due to the SKIP pattern is adjudicated in advance, inter-frame forecast mode coding number of times reduces; The second, enough accurately in situation, no longer do complicated frame mode coding at inter-frame forecast mode.
Table 3 WM3.3a algorithm and algorithm performance index of the present invention contrast
。
Table 3 has been listed the algorithm performance comparative result of algorithm of the present invention and reference model, and experimental result shows, the coding rate of algorithm of the present invention be 2.10～7.04 times of the coding mode selection algorithm of WM3.3a (
QP=40 o'clock, calculate by the scramble time).
QPLarger, in the lower situation of the motion complexity of sequence, effect is wanted significantly, this be because quantization parameter when large in corresponding coding SKIP macro block ratio increase, so improved search speed; The search speed of algorithm of the present invention is along with the complexity of background, and the motion severe degree increases and reduces, because background is more complicated, motion Shaoxing opera is strong, and in corresponding coding, SKIP macro block ratio is just less.For picture quality, adopt as can be seen from Table 3 the predictive mode fast selection algorithm to descend seldom than the PSNR of former algorithm, mean P SNR drops to 0.1dB, and for the larger sequence PSNR of the motion 0.162dB that descends at most, the variation of bit rate is very little.
Should be understood that; above-mentioned explanation is not to be limitation of the present invention, and the present invention also is not limited in above-mentioned giving an example, the modification that those skilled in the art make in essential scope of the present invention; distortion, interpolation or replacement also should belong to protection scope of the present invention.
Claims (8)
1. AVS-M Video coding fast intraframe predicting mode selecting method, its feature comprises:
(a) according to the interframe encoding mode distribution situation, it is classified, arrange by processing priority, adjudicate dissimilar in order, selectively relatively may pattern, effectively reduce the quantity that needs decision pattern, determine fast optimization model;
(b) adopt the premature termination judgement, the SKIP pattern is preferentially adjudicated, whether other pattern passing threshold adjudicates premature termination, if satisfy the threshold value judgment condition, judges that directly this macro block is with the SKIP pattern-coding; If do not satisfy the threshold value judgment condition, with Inter4 * 4, Inter8 * 8, Interl6 * 16 3 kind of pattern is as originate mode, whether meet monotonicity according to the rate distortion costs of these three kinds of patterns, first get rid of a part of macro block mode, reduce useless block mode search procedure, if rate distortion costs does not meet monotonicity, the block mode of search overall dimension.
2. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1, it is characterized in that: according to the interframe encoding mode distribution situation with its method of classifying be: the SKIP pattern is divided into separately a class; Interl6 * 16, Interl6 * 8, Inter8 * 16 are divided into the macro block class; Inter8 * 8, Inter8 * 4, Inter4 * 8, Inter4 * 4 are divided into the sub-block class; Frame mode is as a class.
3. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1, it is characterized in that: the order of arranging by processing priority is: { SKIP}, { Interl6 * 16, Interl6 * 8, Inter8 * 16 }, { Inter8 * 8, Inter8 * 4, Inter4 * 8, Inter4 * 4 }, { Intra-Direct, Intra-4 * 4 }.
4. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1, is characterized in that: threshold value
THDetermine according to full zero-detection threshold value in AVS-M:
5. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1 is characterized in that: if
J Mode(16 * 16)＞
J Mode(8 * 8)＞
J Mode(4 * 4) or
J Mode(16 * 16)＜
J Mode(8 * 8)＜
J Mode(4 * 4), rate distortion costs is dull, if rate distortion costs has monotonicity, only needs the pattern between two kinds of optimal modes of check, for example: 8 * 8,4 * 4 is optimal mode, only need detect 8 * 4 and 4 * 8 patterns again; 16 * 16,8 * 8 is optimal mode, only need detect 16 * 8 and 8 * 16 patterns again.
6. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1, is characterized in that: as follows to the decision rule that inter mode decision is taked: at first search for 16 * 16 and 8 * 8, calculate
J Mode(16 * 16) and
J Mode(8 * 8), if
J Mode(SKIP) simultaneously less than
J Mode(16 * 16) and
J Mode(8 * 8), account for motion are estimated and are compensated the effect that rear its match condition is difficult to surpass the SKIP pattern, select the SKIP pattern as the forced coding pattern, and stop search; Otherwise search for 4 * 4, calculate
J Mode(4 * 4), relatively
J Mode(16 * 16),
J Mode(8 * 8) and
J ModeThe dullness of (4 * 4); If
J Mode(16 * 16) maximum continues search 8 * 4 and 4 * 8, and calculates respectively
J Mode(8 * 4) and
J Mode(4 * 8), relatively
J Mode(8 * 4),
J Mode(4 * 8) and
J Mode(4 * 4); If
J Mode(4 * 4) maximum continues search 16 * 8 and 8 * 16, and calculates respectively
J Mode(16 * 8) and
J Mode(8 * 16), relatively
J Mode(16 * 16),
J Mode(16 * 8) and
J Mode(8 * 16); If rate distortion costs does not meet monotonicity, the block mode of search overall dimension.
7. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1, it is characterized in that: select the decision rule take as follows to frame mode: according to the time domain spatial correlation of video image, with the optimum rate distortion costs of adjacent macroblocks, one decision threshold is set
TIf,
J Mode＞
TCalculate frame mode; Otherwise do not calculate;
J Mode(inter) be minimum rate distortion costs after inter prediction,
TBe defined as follows:
X
0Be current coding macro block, X
1For in former frame with X
0The macro block of same position,
J Mode(A
0),
J Mode(B
0),
J Mode(C
0),
J Mode(X
1) represent respectively A
0, B
0, C
0And X
1Optimum rate distortion costs:
8. a kind of AVS-M Video coding fast intraframe predicting mode selecting method according to claim 1, it is characterized in that: fast interframe mode selection method comprises the following steps:
Step 1 detects SKIP, calculates
SAD(SKIP);
Step 2
SAD(SKIP)＜
TH, SKIP is optimal mode, forwards step 7 to, otherwise, search 16 * 16 and 8 * 8;
Step 3
J Modo(SKIP)＜
J Modo(16 * 16),
J Modo(SKIP)＜
J Modo(8 * 8), SKIP is optimal mode, forwards step 7 to, otherwise, search 4 * 4;
Step 4
J Modo(16 * 16)＞
J Modo(8 * 8)＞
J ModoWhen (4 * 4), search 8 * 4 and 4 * 8 forwards step 5 to,
J Modo(16 * 16)＜
J Modo(8 * 8)＜
J ModoWhen (4 * 4), search 8 * 4 and 4 * 8,16 * 8 and 8 * 16 forwards step 5 to; Otherwise, search 16 * 8 and 8 * 16;
Step 5 is determined best inter mode;
Step 6
J Modo(inter)＞
T, detect frame mode, determine optimal mode, otherwise, forward step 7 to;
Step 7 end mode selection course.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100569696A CN103384325A (en) | 2013-02-22 | 2013-02-22 | Quick inter-frame prediction mode selection method for AVS-M video coding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100569696A CN103384325A (en) | 2013-02-22 | 2013-02-22 | Quick inter-frame prediction mode selection method for AVS-M video coding |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103384325A true CN103384325A (en) | 2013-11-06 |
Family
ID=49492012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100569696A Pending CN103384325A (en) | 2013-02-22 | 2013-02-22 | Quick inter-frame prediction mode selection method for AVS-M video coding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103384325A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268549A (en) * | 2014-09-16 | 2015-01-07 | 天津大学 | Anti-noise multi-scale local binary pattern characteristic representation method |
| CN105282557A (en) * | 2015-09-28 | 2016-01-27 | 天津大学 | H264 rapid movement estimation method for prediction movement vector |
| CN105578195A (en) * | 2015-12-24 | 2016-05-11 | 福州瑞芯微电子股份有限公司 | H.264 inter-frame prediction system |
| CN105578197A (en) * | 2015-12-24 | 2016-05-11 | 福州瑞芯微电子股份有限公司 | Master control system for realizing inter-frame prediction |
| CN105828084A (en) * | 2016-03-30 | 2016-08-03 | 腾讯科技（深圳）有限公司 | HEVC inter-frame encoding processing method and device |
| CN106686383A (en) * | 2017-01-17 | 2017-05-17 | 湖南优象科技有限公司 | Depth map intra-frame coding method capable of preserving edge of depth map |
| CN107318016A (en) * | 2017-05-08 | 2017-11-03 | 上海大学 | A kind of HEVC inter-frame forecast mode method for rapidly judging based on zero piece of distribution |
| CN107343198A (en) * | 2017-05-08 | 2017-11-10 | 上海大学 | A kind of quick decision method of AVS2 inter-frame forecast modes |
| CN107409217A (en) * | 2015-03-06 | 2017-11-28 | 高通股份有限公司 | Adaptive model Inspection Order for Video coding |
| CN110022477A (en) * | 2019-03-29 | 2019-07-16 | 中南大学 | A kind of inter-frame forecast mode fast selecting method based on CUTree |
| CN110113601A (en) * | 2019-04-11 | 2019-08-09 | 中南大学 | Fast algorithm selection method in HEVC frame based on video pictures textural characteristics |
| CN110430427A (en) * | 2019-08-07 | 2019-11-08 | 杭州微帧信息科技有限公司 | A kind of Fast video coding decision-making technique and system based on SKIP pattern-coding result |
| CN110958443A (en) * | 2019-12-16 | 2020-04-03 | 宁波大学 | A fast coding method for 360-degree video frames |
| 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 |
| CN113542738A (en) * | 2021-09-17 | 2021-10-22 | 杭州微帧信息科技有限公司 | A Method for Fast Decision-Making of Video Coding Modes |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1589023A (en) * | 2004-08-06 | 2005-03-02 | 联合信源数字音视频技术（北京）有限公司 | Coding and decoding method and device for multiple coded list lengthening based on context |
| CN101677405A (en) * | 2008-09-19 | 2010-03-24 | 香港科技大学 | Transcoding-based robust streaming media techniques and systems for compressed video |
-
2013
- 2013-02-22 CN CN2013100569696A patent/CN103384325A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1589023A (en) * | 2004-08-06 | 2005-03-02 | 联合信源数字音视频技术（北京）有限公司 | Coding and decoding method and device for multiple coded list lengthening based on context |
| CN101677405A (en) * | 2008-09-19 | 2010-03-24 | 香港科技大学 | Transcoding-based robust streaming media techniques and systems for compressed video |
Non-Patent Citations (1)
| Title |
|---|
| 张新安，宫彦军，李小武: "一种AVS-M帧间预测模式快速选择算法", 《小型微型计算机系统》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268549A (en) * | 2014-09-16 | 2015-01-07 | 天津大学 | Anti-noise multi-scale local binary pattern characteristic representation method |
| CN107409217B (en) * | 2015-03-06 | 2020-02-21 | 高通股份有限公司 | Method, device and equipment for coding video data |
| CN107409217A (en) * | 2015-03-06 | 2017-11-28 | 高通股份有限公司 | Adaptive model Inspection Order for Video coding |
| CN105282557A (en) * | 2015-09-28 | 2016-01-27 | 天津大学 | H264 rapid movement estimation method for prediction movement vector |
| CN105282557B (en) * | 2015-09-28 | 2018-05-18 | 天津大学 | A kind of H.264 rapid motion estimating method of predicted motion vector |
| CN105578195A (en) * | 2015-12-24 | 2016-05-11 | 福州瑞芯微电子股份有限公司 | H.264 inter-frame prediction system |
| CN105578197A (en) * | 2015-12-24 | 2016-05-11 | 福州瑞芯微电子股份有限公司 | Master control system for realizing inter-frame prediction |
| CN105578195B (en) * | 2015-12-24 | 2019-03-12 | 福州瑞芯微电子股份有限公司 | A kind of H.264 inter-frame prediction system |
| CN105578197B (en) * | 2015-12-24 | 2019-04-30 | 福州瑞芯微电子股份有限公司 | A kind of realization inter-prediction master control system |
| CN105828084A (en) * | 2016-03-30 | 2016-08-03 | 腾讯科技（深圳）有限公司 | HEVC inter-frame encoding processing method and device |
| CN105828084B (en) * | 2016-03-30 | 2021-04-13 | 腾讯科技（深圳）有限公司 | HEVC (high efficiency video coding) inter-frame coding processing method and device |
| CN106686383A (en) * | 2017-01-17 | 2017-05-17 | 湖南优象科技有限公司 | Depth map intra-frame coding method capable of preserving edge of depth map |
| CN107318016A (en) * | 2017-05-08 | 2017-11-03 | 上海大学 | A kind of HEVC inter-frame forecast mode method for rapidly judging based on zero piece of distribution |
| CN107343198A (en) * | 2017-05-08 | 2017-11-10 | 上海大学 | A kind of quick decision method of AVS2 inter-frame forecast modes |
| CN110022477A (en) * | 2019-03-29 | 2019-07-16 | 中南大学 | A kind of inter-frame forecast mode fast selecting method based on CUTree |
| CN110022477B (en) * | 2019-03-29 | 2021-02-19 | 中南大学 | A fast selection method of inter prediction mode based on CUTree |
| CN110113601B (en) * | 2019-04-11 | 2023-01-06 | 中南大学 | HEVC intra-frame rapid algorithm selection method based on video picture texture features |
| CN110113601A (en) * | 2019-04-11 | 2019-08-09 | 中南大学 | Fast algorithm selection method in HEVC frame based on video pictures textural characteristics |
| CN110430427A (en) * | 2019-08-07 | 2019-11-08 | 杭州微帧信息科技有限公司 | A kind of Fast video coding decision-making technique and system based on SKIP pattern-coding result |
| CN110430427B (en) * | 2019-08-07 | 2021-08-17 | 杭州微帧信息科技有限公司 | A fast video coding decision method and system based on SKIP mode coding results |
| 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 |
| CN112839224B (en) * | 2019-11-22 | 2023-10-10 | 腾讯科技（深圳）有限公司 | Prediction mode selection method and device, video coding equipment and storage medium |
| CN110958443A (en) * | 2019-12-16 | 2020-04-03 | 宁波大学 | A fast coding method for 360-degree video frames |
| CN110958443B (en) * | 2019-12-16 | 2021-06-29 | 宁波大学 | A fast coding method for 360-degree video frames |
| CN113542738B (en) * | 2021-09-17 | 2022-04-12 | 杭州微帧信息科技有限公司 | Method for fast decision of video coding mode |
| CN113542738A (en) * | 2021-09-17 | 2021-10-22 | 杭州微帧信息科技有限公司 | A Method for Fast Decision-Making of Video Coding Modes |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103384325A (en) | Quick inter-frame prediction mode selection method for AVS-M video coding | |
| CN107147911B (en) | Method and device for fast inter-frame coding mode selection based on local luminance compensation LIC | |
| CN102984521B (en) | High-efficiency video coding inter-frame mode judging method based on temporal relativity | |
| CN102186070B (en) | Method for realizing rapid video coding by adopting hierarchical structure anticipation | |
| CN104539962B (en) | It is a kind of merge visually-perceptible feature can scalable video coding method | |
| CN103188496B (en) | Based on the method for coding quick movement estimation video of motion vector distribution prediction | |
| CN101217663B (en) | A quick selecting method of the encode mode of image pixel block for the encoder | |
| CN104796694B (en) | Optimization intraframe video coding method based on video texture information | |
| CN101448159B (en) | A fast inter-frame mode selection method based on rate-distortion cost and mode frequency | |
| CN103997646A (en) | Rapid intra-frame prediction mode selection method in high-definition video coding | |
| CN104320658B (en) | A kind of HEVC fast encoding methods | |
| CN103384326A (en) | Quick intra-frame prediction mode selection method for AVS-M video coding | |
| CN102647598B (en) | H.264 inter-frame mode optimization method based on maximum and minimum MV difference | |
| CN101888546B (en) | A kind of method of estimation and device | |
| CN103248895A (en) | Quick mode estimation method used for HEVC intra-frame coding | |
| CN101583036A (en) | Method for determining the relation between movement characteristics and high efficient coding mode in pixel-domain video transcoding | |
| CN103384327A (en) | AVS fast mode selection algorithm based on adaptive threshold | |
| CN101754022A (en) | Motion estimation method with low complexity | |
| CN102075757B (en) | Video foreground object coding method by taking boundary detection as motion estimation reference | |
| CN101867818B (en) | Selection method and device of macroblock mode | |
| CN106791828A (en) | High performance video code-transferring method and its transcoder based on machine learning | |
| CN101072355B (en) | A Weighted Predictive Motion Compensation Method | |
| CN102685497A (en) | Rapid interframe mode selection method and device for AVS (Advanced Audio Video Coding Standard) coder | |
| CN105282557B (en) | A kind of H.264 rapid motion estimating method of predicted motion vector | |
| CN101778281A (en) | H.264 rapid motion estimation method based on structural similarity |
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: 20131106 |