CN101202915A - Method and apparatus for selecting frame inner forecasting mode - Google Patents
Method and apparatus for selecting frame inner forecasting mode Download PDFInfo
- Publication number
- CN101202915A CN101202915A CN 200710124902 CN200710124902A CN101202915A CN 101202915 A CN101202915 A CN 101202915A CN 200710124902 CN200710124902 CN 200710124902 CN 200710124902 A CN200710124902 A CN 200710124902A CN 101202915 A CN101202915 A CN 101202915A
- Authority
- CN
- China
- Prior art keywords
- mode
- predictive mode
- functional value
- cost function
- prediction mode
- 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)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention discloses a selecting method and a device for an intra-prediction mode. The method comprises the following steps: A. establishing the sequence relationships of each prediction model under different most possible prediction modes; B. selecting reference image blocks for the image blocks to be predicted; getting the most possible prediction modes of the image blocks to be predicted by referring to the reference image blocks; C. traversing all the prediction modes according to the sequence relationships; calculating the cost function value of each prediction when traversing thereof, and a minimum cost function value is set to be equal with the cost function value when the cost function value is less than a preset cost function threshold in the calculation of the cost function value in the prediction model, and the intra prediction is finished by taking the resent prediction mode as the best prediction mode and the present block as an all-zero block. The invention greatly reduces the calculating complexity of intra-prediction without affecting the compression quality.
Description
Technical field
The present invention relates to video coding technique, relate in particular to a kind of system of selection and device of intra prediction mode.
Background technology
H.264, AVS (Advanced Video and audio Standard, advanced audio/video encoding standard) etc. in the standard, infra-frame prediction is an important techniques that improves compression efficiency, promptly in cataloged procedure, for the current some zones that will encode (being generally a rectangular block) in the video image, search and its most akin and encoded piece are predicted the current piece that will encode with this most akin in video image.And for color video frequency image, also comprise luminance component frame and chromatic component frame in every frame video image, the luminance block intra prediction mode of existing AVS-M standard is referring to shown in Fig. 1 and the table 1, it comprises 9 kinds of intra prediction modes, i.e. Intra_Luma_Down_Left (left side tiltedly down), Intra_Luma_Vertical_Left (vertically taking back), Intra_Luma_Vertical (vertically), Intra_Luma_Vertical_Right (vertically taking over), Intra_Luma_Down_Right (right side tiltedly down), Intra_Luma_Horizontal_Down (level on the lower side), Intra_Luma_Horizontal (level), Intra_Luma_Horizontal_Up (level on the upper side), Intra_Luma_DC (direct current).
Infra-frame prediction should therefrom select optimum prediction mode, H.264 and the existing way of compression standard such as AVS be predicted pixel values and the calculation cost that from 0～8 nine kind of pattern, obtains every kind of pattern successively, be the cost functional value, relatively obtain the cost reckling as optimum prediction mode.
Yet the probability that these nine kinds of patterns take place do not equate, and need travel through this nine kinds of patterns when finding optimal mode at every turn, therefore often expends some unnecessary times.In addition, prior art is whether detect this piece at DCT, after quantizing be complete zero piece, and promptly each pixel value in this piece is 0; If complete zero piece is then at cbp_4x4[b8] corresponding positions be labeled as 0, otherwise be labeled as 1.But some complete zero piece just can detect before DCT, quantification, there is no need to do DCT, quantification, inverse quantization, IDCT.
Table 1
| In the luminance block frame | Title |
| Predictive mode | |
| 0 | A left side tiltedly down |
| 1 | Vertically take back |
| 2 | Vertically |
| 3 | Vertically take over |
| 4 | Right down oblique |
| 5 | Level on the |
| 6 | |
| 7 | |
| 8 | Direct current |
Summary of the invention
Because above-mentioned defective of the prior art, technical problem to be solved by this invention is: provide a kind of and needn't travel through system of selection and the device that nine kinds of patterns just can find the intra prediction mode of optimal mode, thereby under the condition that does not influence compression quality, reduce computational complexity.
In order to solve the problems of the technologies described above, the system of selection of intra prediction mode of the present invention comprises following steps:
A, the ordinal relation of establishing each predictive mode under the different most possible predictive modes;
B, choose reference image block for image block to be predicted; Obtain the most possible predictive mode of image block to be predicted according to reference image block;
C, according to ordinal relation described in the steps A, obtain the order of each predictive mode of the most possible predictive mode correspondence that obtains among the step B, travel through each predictive mode in proper order by this; Whenever go through a predictive mode, calculate the cost functional value under this predictive mode, when this cost functional value during less than default cost function threshold, set minimum cost functional value and equal this cost functional value, and be optimum prediction mode with current predictive mode, current block is complete zero piece, finishes infra-frame prediction, finishes; Otherwise,, change next step over to when this cost functional value during more than or equal to default cost function threshold;
D, when above-mentioned cost functional value during less than described minimum cost functional value, set minimum cost functional value and equal this cost functional value, setting optimum prediction mode is current predictive mode;
E, judge whether predictive mode travels through end, in this way, finish infra-frame prediction, finish with the optimum prediction mode of determining among the step D; Otherwise, return step C.
Wherein, described cost function threshold is chosen from cost function threshold table Tcost[qb], described cost function threshold table Tcost[qb] dimension qb represent quantization parameter, the tabular value under each dimension is represented the cost function threshold under this dimension corresponding quantitative parameter.
Described cost function threshold table Tcost[qb] be specially: Tcost[64]=
2， 2， 2， 2， 3， 3， 3， 4，
4， 4， 5， 6， 7， 8， 8， 9，
10， 11， 12， 13， 15， 16， 17， 19，
21， 23， 25， 27， 30， 32， 35， 39，
42， 46， 50， 55， 60， 65， 71， 78，
85， 93， 101，110，120，131，143，156，
170，185，203，221，241，263，287，313，
341，371，406，443，482，526，574，624}。
The ordinal relation of each predictive mode is according to the establishment of sorting of optimum prediction mode probability distribution order from big to small under the most possible predictive mode of difference under the most possible predictive mode of described difference, and this ordinal relation is stored in most possible pattern-optimal mode matrix.
Described most possible pattern-optimal mode matrix is specially most_to_best_mode[9] [9]=
{0，6，5，4，3，1，2，7，8}，
{1，3，6，0，2，5，4，7，8}，
{2，6，5，0，7，4，3，1，8}，
{3，2，1，4，6，5，0，7，8}，
{4，1，7，6，0，2，3，5，8}，
{5，1，7，6，3，0，2，4，8}，
{6，1，2，5，4，0，7，3，8}，
{7，1，5，3，4，6，0，2，8}，
8,6,7,2,1,5,3,4,0}}; Wherein, rectangular array is represented most possible predictive mode, and row matrix is represented optimum prediction mode.
The system of selection of described intra prediction mode, described reference image block are the adjacent left image block and the adjacent top image block of image block to be predicted.
The choice device of intra prediction mode of the present invention comprises:
Map unit, the ordinal relation that is used to establish each predictive mode under the different most possible predictive modes;
Most possible predictive mode acquiring unit is used to image block to be predicted to choose reference image block and obtains the most possible predictive mode of image block to be predicted according to reference image block;
Predictive mode traversal unit, the ordinal relation that is used for each predictive mode of the most possible predictive mode that obtains according to above-mentioned most possible predictive mode acquiring unit and this most possible predictive mode correspondence of above-mentioned map unit, travel through each predictive mode, whenever go through a predictive mode, calculate the cost functional value under this predictive mode;
The comparison prediction unit, being used for that predictive mode is traveled through unit cost functional value that calculates and the cost function threshold of presetting compares, when this cost functional value during less than default cost function threshold, set minimum cost functional value and equal this cost functional value, and be optimum prediction mode with current predictive mode, current block is complete zero piece, finishes infra-frame prediction;
Judge setup unit, be used for when this cost functional value during more than or equal to default cost function threshold, judge that whether the cost functional value is less than described minimum cost functional value, if above-mentioned cost functional value is less than described minimum cost functional value, then set minimum cost functional value and equal this cost functional value, setting optimum prediction mode is current predictive mode; And
Judge predicting unit, be used to judge whether predictive mode travels through end, in this way, finish infra-frame prediction to judge the optimum prediction mode that setup unit is set; Otherwise control predictive mode traversal unit continues traversal.
The choice device of described intra prediction mode, described cost function threshold is chosen from cost function threshold table Tcost[qb], described cost function threshold table Tcost[qb] dimension qb represent quantization parameter, tabular value under each dimension is represented the cost function threshold under this dimension corresponding quantitative parameter, described cost function threshold table Tcost[qb] be specially: Tcost[64]=
2， 2， 2， 2， 3， 3， 3， 4，
4， 4， 5， 6， 7， 8， 8， 9，
10， 11， 12， 13， 15， 16， 17， 19，
21， 23， 25， 27， 30， 32， 35， 39，
42， 46， 50， 55， 60， 65， 71， 78，
85， 93， 101，110，120，131，143，156，
170，185，203，221，241，263，287，313，
341，371，406，443，482，526，574，624}。
The choice device of described intra prediction mode, the ordinal relation of each predictive mode is to sort according to optimum prediction mode probability distribution order from big to small under the most possible predictive mode of difference to establish under the most possible predictive mode of described difference, and this ordinal relation constitutes most possible pattern-optimal mode matrix.
Described most possible pattern-optimal mode matrix is specially most_to_best_mode[9] [9]=
{0，6，5，4，3，1，2，7，8}，
{1，3，6，0，2，5，4，7，8}，
{2，6，5，0，7，4，3，1，8}，
{3，2，1，4，6，5，0，7，8}，
{4，1，7，6，0，2，3，5，8}，
{5，1，7，6，3，0，2，4，8}，
{6，1，2，5，4，0，7，3，8}，
{7，1，5，3，4，6，0，2，8}，
8,6,7,2,1,5,3,4,0}}; Wherein, rectangular array is represented most possible predictive mode, and row matrix is represented optimum prediction mode.
The present invention passes through to establish the ordinal relation and the setting cost function threshold of each predictive mode under the different most possible patterns, under such setting, there is very big probability to find optimal mode in preceding several patterns, avoid simultaneously complete zero piece is done dct transform, quantification, inverse quantization, inverse transformation, reduce computational complexity, improved the selection speed of the optimum prediction mode of intra prediction mode.
Description of drawings
Fig. 1 is 9 kinds of intra prediction mode schematic diagrames of luminance block in the prior art;
Fig. 2 is the flow chart of the fast selecting method of the specific embodiment of the invention.
Embodiment
Contrast accompanying drawing below, the present invention is described in detail in conjunction with embodiment.
Referring to Fig. 2, in the AVS-M standard, luminance block adopts 4 * 4 pattern usually.The system of selection of intra prediction mode of the present invention at first is that image block to be predicted is selected reference image block, in normal circumstances, reference image block is all chosen adjacent top image block and adjacent left image block, according to the standard code of AVS-M, then can obtain the most possible predictive mode mostProbableMode of image block to be predicted by the pattern of adjacent top image block and adjacent left image block then.
The present invention (is respectively the bus_cif.yuv of CIF form with 4 international graphics standard sequences, football_cif.yuv, foreman_cif.yuv, mobile_cif.yuv) be benchmark, adopt methods such as statistics, statistics is under the certain condition of most possible predictive mode mostProbableMode, different masses is chosen the probability distribution of optimum prediction mode, set up most possible-optimal mode matrix with this probability distribution relation, the concrete form of this matrix is: most_to_best_mode[9] [9]=
{0，6，5，4，3，1，2，7，8}，
{1，3，6，0，2，5，4，7，8}，
{2，6，5，0，7，4，3，1，8}，
{3，2，1，4，6，5，0，7，8}，
{4，1，7，6，0，2，3，5，8}，
{5，1，7，6，3，0，2，4，8}，
{6，1，2，5，4，0，7，3，8}，
{7，1，5，3，4，6，0，2，8}，
{8，6，7，2，1，5，3，4，0}
; Wherein matrix is listed as the 9th row from first and has represented most possible predictive mode from 0 to 8, and row matrix has then reflected the probability distribution of the optimum prediction mode under the different most possible predictive modes.This matrix is not limited to any value.As long as the similar matrix of employing is finished the ordering to pattern, all belong to protection range of the present invention.At different standards, method of the present invention can draw different matrixes.
The most possible certain condition of predictive mode mostProbableMode just is meant that under the condition of same most possible predictive mode (in the above-mentioned general image standard sequence, each sequence has 300 two field pictures to different masses, every two field picture such as CIF form 88 * 72=6336 piece arranged) the optimum prediction mode difference chosen, by adding up 4 international graphics standard sequences, can draw the probability distribution that they choose optimal mode, probability distribution is optimal mode and is respectively 0,1,2,3,4,5,6,7,8 o'clock piece number, for example above-mentioned matrix first row has reflected at most possible predictive mode mostProbableMode to be got 0 o'clock, by the piece number of obtaining optimal mode from many be 0,6,5 successively to few tactic each predictive mode, 4,3,1,2,7,8.That is to say that most possible predictive mode-best matrix has reflected the ordinal relation that each predictive mode is arranged by the probability distribution of optimal mode under the different most possible predictive modes.
Utilize above-mentioned graphics standard sequence equally, adopt methods such as statistics, count under the certain condition of quantization parameter qp, zero piece probability P k (zeroblock|cost=k) can obtain cost function threshold table Tcost[qb under the condition of cost=k]: Tcost[64]=
2， 2， 2， 2， 3， 3， 3， 4，
4， 4， 5， 6， 7， 8， 8， 9，
10， 11， 12， 13， 15， 16， 17， 19，
21， 23， 25， 27， 30， 32， 35， 39，
42， 46， 50， 55， 60， 65， 71， 78，
85， 93， 101，110，120，131，143，156，
170，185，203，221，241，263，287，313，
341，371，406，443，482，526，574，624}。
Above-mentioned cost function threshold table Tcost[64] in fact represent the one dimension matrix, by this threshold matrix is set, can the appreciable impact compression efficiency in case the cost functional value less than corresponding threshold value, can shift to an earlier date.This matrix is not limited to any value.As long as the similar matrix of employing is finished the premature termination to pattern, all belong to protection range of the present invention.At different standards, method of the present invention can draw different matrixes.The dimension of cost function threshold matrix is being represented quantization parameter qb, and 64 dimensions are quantization parameter from 0 to 63.Each tabular value reflected under current dimension corresponding quantitative parameter, and when cost functional value during less than this tabular value, image block to be predicted is complete zero piece.For example numerical value 2 expressions of first trip in first is that quantization parameter qb got 0 o'clock, and like this, we can select suitable quantization parameter, qb=40 for example, and with the corresponding Tcost value of this value, promptly 85 as the cost function threshold then.Thus, carry out infra-frame prediction:
1, according to last piece and left block mode determine the most possible predictive mode of current block;
2, according to most possible pattern-optimal mode matrix, select a predictive mode as current predictive mode by each the predictive mode ordinal relation under this most possible predictive mode;
3, dope the pixel value of current block according to current predictive mode;
4, calculate the cost functional value; The cost functional value for example can adopt following formula to calculate:
cost＝sad+f(qp，mod e)
Wherein,
Lambda obtains by tabling look-up:
lambda(qp)＝QP2QUANT[qp]；
QP2QUANT[64]＝
{
1， 1， 1， 1， 1， 2， 2， 2，
2， 2， 2， 3， 3， 3， 3， 4，
4， 4， 5， 5， 6， 6， 7， 7，
8， 9， 10， 10， 11， 12， 13， 15，
16， 17， 19， 21， 23， 25， 27， 29，
32， 35， 38， 41， 45， 49， 54， 59，
64， 70， 76， 83， 91， 99， 108，117，
128，140，152，166，181，197，215，235，
}；
Sad obtains by following formula:
In AVS, M=N=3, f
k(m, n) (m, n) pixel value, the f in the expression current block
K-1(m+i, n+j) value of the corresponding pixel points of expression prediction piece.
5, the cost function threshold with cost functional value and aforementioned setting compares, and when cost functional value during less than the cost function threshold, carry out step 6; Otherwise, jump to step 7;
6, set minimum cost functional value mincost and equal this cost functional value, and be optimum prediction mode with current predictive mode, current block is complete zero piece, is the present mode predicted pixel values with the predicted pixel values of optimal mode, the reconstruction value of infra-frame prediction is the predicted pixel values of optimal mode, finishes;
7, compare cost functional value and minimum cost functional value mincost, if the cost functional value carry out step 8 less than minimum cost functional value mincost; Otherwise, jump to step 9;
8, make minimum cost functional value mincost equal this cost functional value, optimum prediction mode is a present mode, carries out next step;
9, press the ordinal relation of each predictive mode under this most possible predictive mode in most possible pattern-optimal mode matrix, select next predictive mode under the current predictive mode, promptly each predictive mode under this most possible predictive mode in most possible pattern-optimal mode matrix is traveled through, if do not traveled through each predictive mode, then turn back to step 2 and continue traversal; Otherwise, carry out next step;
10, be optimum prediction mode with the predictive mode of determining in the step 8, carry out dct, quantification, inverse quantization, idct, infra-frame prediction is finished in reconstruct.
As shown in Figure 3, the invention also discloses a kind of choice device of intra prediction mode, comprise:
Map unit, the ordinal relation that is used to establish each predictive mode under the different most possible predictive modes;
Most possible predictive mode acquiring unit is used to image block to be predicted to choose reference image block and obtains the most possible predictive mode of image block to be predicted according to reference image block;
Predictive mode traversal unit, the ordinal relation that is used for each predictive mode of the most possible predictive mode that obtains according to above-mentioned most possible predictive mode acquiring unit and this most possible predictive mode correspondence of above-mentioned map unit, travel through each predictive mode, whenever go through a predictive mode, calculate the cost functional value under this predictive mode;
The comparison prediction unit, being used for that predictive mode is traveled through unit cost functional value that calculates and the cost function threshold of presetting compares, when this cost functional value during less than default cost function threshold, set minimum cost functional value and equal this cost functional value, and be optimum prediction mode with current predictive mode, current block is complete zero piece, finishes infra-frame prediction;
Judge setup unit, be used for when this cost functional value during more than or equal to default cost function threshold, judge that whether the cost functional value is less than described minimum cost functional value, if above-mentioned cost functional value is less than described minimum cost functional value, then set minimum cost functional value and equal this cost functional value, setting optimum prediction mode is current predictive mode; And
Judge predicting unit, be used to judge whether predictive mode travels through end, in this way, finish infra-frame prediction to judge the optimum prediction mode that setup unit is set; Otherwise control predictive mode traversal unit continues traversal.
Described choice device is operated according to aforementioned system of selection of the present invention, repeats no more.
The present invention because the probability rearrangement according to condition of mode cycle order and provide the threshold value Tcost[qp of circulation premature termination], under such setting, the preferential predictive mode of selecting optimum prediction mode possibility maximum when at every turn selecting predictive mode, there is very big probability to find optimal mode to jump out circulation in preceding several patterns, avoid simultaneously complete zero piece is done dct transform, quantification, inverse quantization, inverse transformation, reduce computational complexity, improved the selection speed of the optimum prediction mode of intra prediction mode.
Above-mentioned explanation is an example with AVS-M, is appreciated that the present invention protects to be not limited only to AVS-M, is equally applicable to the algorithm of other any employing infra-frame prediction
Above content be in conjunction with concrete preferred implementation to further describing that the present invention did, can not assert that concrete enforcement of the present invention is confined to these explanations.For the general technical staff of the technical field of the invention, under the premise of not departing from the present invention, can also make some simple deduction or replace, all should be considered as belonging to protection scope of the present invention.
Claims (10)
1. the system of selection of an intra prediction mode is characterized in that, comprises following steps:
A, the ordinal relation of establishing each predictive mode under the different most possible predictive modes;
B, choose reference image block for image block to be predicted; Obtain the most possible predictive mode of image block to be predicted according to reference image block;
C, according to ordinal relation described in the steps A, obtain the order of each predictive mode of the most possible predictive mode correspondence that obtains among the step B, travel through each predictive mode in proper order by this; Whenever go through a predictive mode, calculate the cost functional value under this predictive mode, when this cost functional value during less than default cost function threshold, set minimum cost functional value and equal this cost functional value, and be optimum prediction mode with current predictive mode, current block is complete zero piece, finishes infra-frame prediction, finishes; Otherwise,, change next step over to when this cost functional value during more than or equal to default cost function threshold;
D, when above-mentioned cost functional value during less than described minimum cost functional value, set minimum cost functional value and equal this cost functional value, setting optimum prediction mode is current predictive mode;
E, judge whether predictive mode travels through end, in this way, finish infra-frame prediction, finish with the optimum prediction mode of determining among the step D; Otherwise, return step C.
2. the system of selection of intra prediction mode according to claim 1, it is characterized in that, described cost function threshold is chosen from cost function threshold table Tcost[qb], described cost function threshold table Tcost[qb] dimension qb represent quantization parameter, the tabular value under each dimension is represented the cost function threshold under this dimension corresponding quantitative parameter.
3. the system of selection of intra prediction mode according to claim 2 is characterized in that, described cost function threshold table Tcost[qb] be specially: Tcost[64]=
2， 2， 2， 2， 3， 3， 3， 4，
4， 4， 5， 6， 7， 8， 8， 9，
10， 11， 12， 13， 15， 16， 17， 19，
21， 23， 25， 27， 30， 32， 35， 39，
42， 46， 50， 55， 60， 65， 71， 78，
85， 93， 101，110，120，131，143，156，
170，185，203，221，241，263，287，313，
341，371，406，443，482，526，574，624}。
4. the system of selection of intra prediction mode according to claim 1, it is characterized in that, the ordinal relation of each predictive mode is according to the establishment of sorting of optimum prediction mode probability distribution order from big to small under the most possible predictive mode of difference under the most possible predictive mode of described difference, and this ordinal relation is stored in most possible pattern-optimal mode matrix.
5. the system of selection of intra prediction mode according to claim 4 is characterized in that, described most possible pattern-optimal mode matrix is specially most_to_best_mode[9] [9]=
{0，6，5，4，3，1，2，7，8}，
{1，3，6，0，2，5，4，7，8}，
{2，6，5，0，7，4，3，1，8}，
{3，2，1，4，6，5，0，7，8}，
{4，1，7，6，0，2，3，5，8}，
{5，1，7，6，3，0，2，4，8}，
{6，1，2，5，4，0，7，3，8}，
{7，1，5，3，4，6，0，2，8}，
{8，6，7，2，1，5，3，4，0}
; Wherein, rectangular array is represented most possible predictive mode, and row matrix is represented optimum prediction mode.
6. according to the system of selection of each described intra prediction mode of claim 1 to 5, it is characterized in that described reference image block is the adjacent left image block and the adjacent top image block of image block to be predicted.
7. the choice device of an intra prediction mode is characterized in that, comprises:
Map unit, the ordinal relation that is used to establish each predictive mode under the different most possible predictive modes;
Most possible predictive mode acquiring unit is used to image block to be predicted to choose reference image block and obtains the most possible predictive mode of image block to be predicted according to reference image block;
Predictive mode traversal unit, the ordinal relation that is used for each predictive mode of the most possible predictive mode that obtains according to above-mentioned most possible predictive mode acquiring unit and this most possible predictive mode correspondence of above-mentioned map unit, travel through each predictive mode, whenever go through a predictive mode, calculate the cost functional value under this predictive mode;
The comparison prediction unit, being used for that predictive mode is traveled through unit cost functional value that calculates and the cost function threshold of presetting compares, when this cost functional value during less than default cost function threshold, set minimum cost functional value and equal this cost functional value, and be optimum prediction mode with current predictive mode, current block is complete zero piece, finishes infra-frame prediction;
Judge setup unit, be used for when this cost functional value during more than or equal to default cost function threshold, judge that whether the cost functional value is less than described minimum cost functional value, if above-mentioned cost functional value is less than described minimum cost functional value, then set minimum cost functional value and equal this cost functional value, setting optimum prediction mode is current predictive mode; And
Judge predicting unit, be used to judge whether predictive mode travels through end, in this way, finish infra-frame prediction to judge the optimum prediction mode that setup unit is set; Otherwise control predictive mode traversal unit continues traversal.
8. the choice device of intra prediction mode according to claim 7, it is characterized in that, described cost function threshold is chosen from cost function threshold table Tcost[qb], described cost function threshold table Tcost[qb] dimension qb represent quantization parameter, tabular value under each dimension is represented the cost function threshold under this dimension corresponding quantitative parameter, described cost function threshold table Tcost[qb] be specially: Tcost[64]=
2， 2， 2， 2， 3， 3， 3， 4，
4， 4， 5， 6， 7， 8， 8， 9，
10， 11， 12， 13， 15， 16， 17， 19，
21， 23， 25， 27， 30， 32， 35， 39，
42， 46， 50， 55， 60， 65， 71， 78，
85， 93， 101，110，120，131，143，156，
170，185，203，221，241，263，287，313，
341，371，406，443，482，526，574，624}。
9. the choice device of intra prediction mode according to claim 7, it is characterized in that, the ordinal relation of each predictive mode is to sort according to optimum prediction mode probability distribution order from big to small under the most possible predictive mode of difference to establish under the most possible predictive mode of described difference, and this ordinal relation constitutes most possible pattern-optimal mode matrix.
10. the choice device of intra prediction mode according to claim 9 is characterized in that, described most possible pattern-optimal mode matrix is specially most_to_best_mode[9] [9]=
{0，6，5，4，3，1，2，7，8}，
{1，3，6，0，2，5，4，7，8}，
{2，6，5，0，7，4，3，1，8}，
{3，2，1，4，6，5，0，7，8}，
{4，1，7，6，0，2，3，5，8}，
{5，1，7，6，3，0，2，4，8}，
{6，1，2，5，4，0，7，3，8}，
{7，1，5，3，4，6，0，2，8}，
{8，6，7，2，1，5，3，4，0}
; Wherein, rectangular array is represented most possible predictive mode, and row matrix is represented optimum prediction mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710124902 CN101202915A (en) | 2007-12-06 | 2007-12-06 | Method and apparatus for selecting frame inner forecasting mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710124902 CN101202915A (en) | 2007-12-06 | 2007-12-06 | Method and apparatus for selecting frame inner forecasting mode |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101202915A true CN101202915A (en) | 2008-06-18 |
Family
ID=39517841
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200710124902 Pending CN101202915A (en) | 2007-12-06 | 2007-12-06 | Method and apparatus for selecting frame inner forecasting mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101202915A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101854551A (en) * | 2010-06-08 | 2010-10-06 | 浙江大学 | Intra-frame prediction mode coding and decoding method and device |
| CN102300088A (en) * | 2010-06-25 | 2011-12-28 | 财团法人工业技术研究院 | In-frame prediction mode optimization method as well as image compression method and device |
| CN104639940A (en) * | 2015-03-06 | 2015-05-20 | 宁波大学 | Quick HEVC (High Efficiency Video Coding) inter-frame prediction mode selection method |
| CN104796701A (en) * | 2015-03-27 | 2015-07-22 | 北京君正集成电路股份有限公司 | Predication mode determination method and device based on HEVC (High Efficiency Video Coding) |
| CN107197280A (en) * | 2011-06-17 | 2017-09-22 | Jvc 建伍株式会社 | Picture coding device, method for encoding images and recording medium |
| CN110602497A (en) * | 2011-01-24 | 2019-12-20 | 索尼公司 | Image decoding device, image decoding method, and non-transitory computer readable medium |
| WO2020062161A1 (en) * | 2018-09-29 | 2020-04-02 | 富士通株式会社 | Method and apparatus for determining most probable modes for intra-frame prediction, and electronic device |
| CN113347436A (en) * | 2019-06-21 | 2021-09-03 | 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 |
-
2007
- 2007-12-06 CN CN 200710124902 patent/CN101202915A/en active Pending
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101854551A (en) * | 2010-06-08 | 2010-10-06 | 浙江大学 | Intra-frame prediction mode coding and decoding method and device |
| CN101854551B (en) * | 2010-06-08 | 2012-08-15 | 浙江大学 | Intra-frame prediction mode coding and decoding method and device |
| CN102300088A (en) * | 2010-06-25 | 2011-12-28 | 财团法人工业技术研究院 | In-frame prediction mode optimization method as well as image compression method and device |
| CN102300088B (en) * | 2010-06-25 | 2013-11-06 | 财团法人工业技术研究院 | In-frame prediction mode optimization method as well as image compression method and device |
| CN110602497A (en) * | 2011-01-24 | 2019-12-20 | 索尼公司 | Image decoding device, image decoding method, and non-transitory computer readable medium |
| CN110602497B (en) * | 2011-01-24 | 2021-11-30 | 索尼公司 | Image decoding device, image decoding method, and non-transitory computer readable medium |
| CN107197280A (en) * | 2011-06-17 | 2017-09-22 | Jvc 建伍株式会社 | Picture coding device, method for encoding images and recording medium |
| CN107197280B (en) * | 2011-06-17 | 2020-06-16 | Jvc 建伍株式会社 | Image encoding device, image encoding method, and recording medium |
| CN104639940B (en) * | 2015-03-06 | 2017-10-10 | 宁波大学 | A kind of quick HEVC method for choosing frame inner forecast mode |
| CN104639940A (en) * | 2015-03-06 | 2015-05-20 | 宁波大学 | Quick HEVC (High Efficiency Video Coding) inter-frame prediction mode selection method |
| CN104796701A (en) * | 2015-03-27 | 2015-07-22 | 北京君正集成电路股份有限公司 | Predication mode determination method and device based on HEVC (High Efficiency Video Coding) |
| CN104796701B (en) * | 2015-03-27 | 2018-02-16 | 北京君正集成电路股份有限公司 | Predictive mode based on HEVC determines method and device |
| WO2020062161A1 (en) * | 2018-09-29 | 2020-04-02 | 富士通株式会社 | Method and apparatus for determining most probable modes for intra-frame prediction, and electronic device |
| CN113347436A (en) * | 2019-06-21 | 2021-09-03 | 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 |
| CN113347436B (en) * | 2019-06-21 | 2022-03-08 | 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101202915A (en) | Method and apparatus for selecting frame inner forecasting mode | |
| CN101964906B (en) | Rapid intra-frame prediction method and device based on texture characteristics | |
| CN102196256B (en) | Video coding method and device | |
| CN101217663B (en) | A quick selecting method of the encode mode of image pixel block for the encoder | |
| CN103546749B (en) | Method for optimizing HEVC (high efficiency video coding) residual coding by using residual coefficient distribution features and bayes theorem | |
| CN102065298B (en) | High-performance macroblock coding implementation method | |
| CN105120292A (en) | Video coding intra-frame prediction method based on image texture features | |
| CN102025995B (en) | Spatial enhancement layer rapid mode selection method of scalable video coding | |
| CN100574447C (en) | Fast intraframe predicting mode selecting method based on the AVS video coding | |
| CN103188496A (en) | Fast motion estimation video encoding method based on motion vector distribution forecast | |
| CN106937116A (en) | Low-complexity video coding method based on random training set adaptive learning | |
| CN101056408B (en) | Method and apparatus for determining a threshold for skip mode in video coding | |
| CN102196272B (en) | P frame coding method and device | |
| CN101742288B (en) | Video noise reduction encoding method and video noise reduction encoding device | |
| EP2661079A1 (en) | H264 transcoding method by multiplexing code stream information | |
| CN105187826A (en) | Rapid intra-frame mode decision method specific to high efficiency video coding standard | |
| CN103888763B (en) | Intra-frame coding method based on HEVC | |
| CN104113761B (en) | Bit rate control method and encoder in a kind of Video coding | |
| CN101883275B (en) | Video coding method | |
| CN101304529A (en) | Method and device for selecting macro block pattern | |
| CN1613259A (en) | Method and system for detecting intra-coded pictures and for extracting intra DCT precision and macroblock-level coding parameters from uncompressed digital video | |
| CN101742301B (en) | Block mode coding method and device | |
| CN101742290A (en) | Methods and devices for denoising in video coding and decoding, | |
| CN104333755B (en) | The CU based on SKIP/Merge RD Cost of B frames shifts to an earlier date terminating method in HEVC | |
| CN101426140B (en) | AVS fast schema selection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20080618 |