CN104754338A - Selection method and device for intra-frame predication mode - Google Patents
Selection method and device for intra-frame predication mode Download PDFInfo
- Publication number
- CN104754338A CN104754338A CN201410850621.9A CN201410850621A CN104754338A CN 104754338 A CN104754338 A CN 104754338A CN 201410850621 A CN201410850621 A CN 201410850621A CN 104754338 A CN104754338 A CN 104754338A
- Authority
- CN
- China
- Prior art keywords
- unit
- predictive mode
- cost
- sampling
- preliminary election
- 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
Abstract
The embodiment of the invention provides a selection method and device for an intra-frame predication mode. The selection method for the intra-frame predication mode particularly comprises the following steps: carrying out lower sampling on an encoding unit to obtain a corresponding sampling unit; predicating the sampling unit by using one of a plurality of predication modes to obtain a corresponding predication unit; calculating a predication cost between each predication unit and the encoding unit; selecting the predication mode from the plurality of predication modes according to the predication cost of each predication unit; and calculating a corresponding rate-distortion cost aiming at the pre-selected predication mode, and selecting the optimal predication mode from the pre-selected predication mode according to the rate-distortion cost. According to the selection method, the complexity of an intra-frame predication mode selection algorithm can be reduced and the encoding efficiency is improved.
Description
Technical field
The present invention relates to technical field of video coding, particularly relate to a kind of system of selection and device of intra prediction mode.
Background technology
HEVC (High Efficiency Video Coding, efficient video coding standard) is video encoding standard of new generation, and its object produced is the demand that solution people increase day by day to vision and acoustical quality.
For video encoding standard, an important part is predictive coding.The quality of prediction directly determines the effect of Video coding, and the selection of predictive mode occupies sizable scramble time, and thus, how can select best pattern fast and accurately from all predictive modes is a study hotspot always.
Compared to kind of the intra prediction mode of 9 H.264, in order to reach better coding efficiency, HEVC standard proposes 35 kinds of predictive modes, is intended to the precision improving infra-frame prediction.But more predictive mode brings higher computation complexity equally.
In HEVC, one two field picture is divided into LCU (the Largest CodingUnit of many non-overlapping copies, maximum coding unit), each LCU is of a size of 64 × 64, each LCU can be many CU (Coding Unit by the model split of quaternary tree recurrence again, coding unit), when LCU does not divide, it is of a size of the CU of 64 × 64, the degree of depth (depth) is defined as 0, when LCU is divided into 4 sub-CU, it is 1 that every sub-CU is of a size of 32 × 32, depth, and every sub-CU can divide independent of other CU recurrence.Group CU is of a size of 8 × 8, when namely depth is 3, no longer continues to divide.
HEVC is exactly first carry out intra prediction mode search to LCU to the process that a LCU (depth=0) carries out intraframe coding, and calculate corresponding rate distortion costs RdCost_1, then LCU is divided into 4 sub-CU (depth=1), respectively 35 kinds of predictive mode search are carried out to every sub-CU, and obtain corresponding rate distortion costs.Finally obtain total rate distortion costs RdCost_2 of 4 sub-CU, the relatively value of RdCost_1 and RdCost_2, if the former is little, using the optimal prediction modes of optimum prediction mode corresponding for RdCost_1 as LCU, otherwise using the optimal prediction modes of optimum prediction mode corresponding for RdCost_2 as LCU.Similarly, when intra prediction mode search is carried out to every sub-CU, the process identical with LCU can be experienced, namely also will be divided into 4 less sub-CU (depth=2), and then compare, so divide down until the depth=3 of CU.RdCost_2 be 4 sub-CU cost and, the cost of sub-CU itself is also that level determines, namely can be that it is encoded as a CU, also can continue to be divided into 4 less CU and encode, and the cost of sub-CU selects less that from the two.And the dividing condition of 4 sub-CU is Independent Decisiveness, is not affected each other.Although the infra-frame prediction of exhaustive search improves precision of prediction, also substantially increase encoder complexity and scramble time.
Summary of the invention
Embodiment of the present invention technical problem to be solved is to provide a kind of system of selection and device of intra prediction mode, can reduce the complexity surveying mode selection algorithm in pre-frame, improves code efficiency.
In order to solve the problem, the invention discloses a kind of system of selection of intra prediction mode, comprising:
Down-sampling is carried out to coding unit, obtains corresponding sampling unit;
Each utilizing in multiple predictive mode is respectively predicted described sampling unit, obtains corresponding predicting unit;
Calculate the prediction cost between each predicting unit and described coding unit;
According to the prediction cost that each predicting unit is corresponding, from described multiple predictive mode, select preliminary election predictive mode;
Calculate corresponding rate distortion costs for described preliminary election predictive mode, and from all preliminary election predictive modes, select best predictive mode according to described rate distortion costs.
Preferably, described down-sampling is carried out to coding unit, obtains corresponding sampling unit, comprising:
The each one times of down-sampling of horizontal and vertical is carried out respectively to coding unit, obtains corresponding sampling unit; Wherein, described sampling unit comprises the pixel of even number line even column or even number line odd column or odd-numbered line odd column or odd-numbered line even column in described coding unit.
Preferably, the prediction cost between each predicting unit of described calculating and described coding unit, comprising:
Calculate between each predicting unit and described coding unit Adama cost parameter;
Prediction cost corresponding to each predicting unit is calculated according to the Adama cost after weighting and code check cost.
Preferably, the described prediction cost corresponding according to each predicting unit, from described multiple predictive mode, select preliminary election predictive mode, comprising:
According to the prediction cost that each predicting unit is corresponding, from described multiple predictive mode select prediction Least-cost one or more as preliminary election predictive mode.
Preferably, the described rate distortion costs calculating correspondence for described preliminary election predictive mode, and from all preliminary election predictive modes, select best predictive mode according to described rate distortion costs, comprising:
Corresponding rate distortion costs is calculated for described preliminary election predictive mode;
Determine that the minimum preliminary election predictive mode of rate distortion costs is best predictive mode.
According to another aspect of the present invention, provide a kind of choice device of intra prediction mode, comprising:
Sampling module, for carrying out down-sampling to coding unit, obtains corresponding sampling unit;
Prediction module, predicts described sampling unit for each utilizing in multiple predictive mode respectively, obtains corresponding predicting unit;
Computing module, for calculating the prediction cost between each predicting unit and described coding unit;
Preliminary election module, for according to prediction cost corresponding to each predicting unit, selects preliminary election predictive mode from described multiple predictive mode; And
Selecting module, for calculating corresponding rate distortion costs for described preliminary election predictive mode, and from all preliminary election predictive modes, selecting best predictive mode according to described rate distortion costs.
Preferably, described sampling module, comprising:
Sampling unit, for carrying out each one times of down-sampling of horizontal and vertical respectively to coding unit, obtains corresponding sampling unit; Wherein, described sampling unit comprises the pixel of even number line even column or even number line odd column or odd-numbered line odd column or odd-numbered line even column in described coding unit.
Preferably, described computing module, comprising:
First computing unit, for calculate between each predicting unit and described coding unit Adama cost parameter;
Second computing unit, for calculating prediction cost corresponding to each predicting unit according to the Adama cost after weighting and code check cost.
Preferably, described preliminary election module, comprising:
Pre-selection unit, for according to prediction cost corresponding to each predicting unit, one or more selecting prediction Least-cost from described multiple predictive mode are as preliminary election predictive mode.
Preferably, described selection module, comprising:
Cost calculating unit, for calculating corresponding rate distortion costs for described preliminary election predictive mode;
Pattern determining unit, for determining that the minimum preliminary election predictive mode of rate distortion costs is best predictive mode.
Compared with prior art, the embodiment of the present invention comprises following advantage:
The embodiment of the present invention carries out down-sampling to coding unit, obtain corresponding sampling unit, and the selection of predictive mode is carried out for sampling unit, pixel due to sampling unit is less than the pixel of coding unit, therefore the calculative data volume of selection algorithm can be simplified, thus the complexity of selection algorithm can be reduced;
In addition, the embodiment of the present invention is not directly for all predictive mode calculation rate distortion costs, but first from described multiple predictive mode, selects preliminary election predictive mode, then for described preliminary election predictive mode calculation rate distortion cost; This, relative to for all predictive mode calculation rate distortion costs, greatly reduces the data volume involved by calculation rate distortion cost, thus can reduce the complexity of selection algorithm further.
Accompanying drawing explanation
Fig. 1 shows the flow chart of steps of the system of selection embodiment of a kind of intra prediction mode of the present invention;
Fig. 2 shows a kind of schematic diagram utilizing 35 kinds of predictive modes to predict block of pixels of the present invention; And
Fig. 3 shows the structured flowchart of the choice device embodiment of a kind of intra prediction mode of the present invention.
Embodiment
For enabling above-mentioned purpose of the present invention, feature and advantage become apparent more, and below in conjunction with the drawings and specific embodiments, the present invention is further detailed explanation.
Embodiment one
With reference to Fig. 1, show the flow chart of steps of the system of selection embodiment of a kind of intra prediction mode of the present invention, specifically can comprise:
Step 101, down-sampling is carried out to coding unit, obtain corresponding sampling unit;
Owing to often there is extremely strong correlation in vedio data, the space length between pixel is nearer, and so their correlation is also stronger.The present invention utilizes this correlation, is simplified by coding unit, and like this, the selection algorithm for predictive mode needs data volume to be processed to greatly reduce, thus can reduce complexity and the operation time of selection algorithm.
The present invention realizes simplifying of coding unit by carrying out down-sampling to coding unit, and like this, the sampling unit corresponding to described coding unit carries out infra-frame prediction, can reduce the workload of selection algorithm.
In one preferred embodiment of the invention, described down-sampling is carried out to coding unit, obtain corresponding sampling unit, specifically can comprise: each one times of down-sampling of horizontal and vertical is carried out respectively to coding unit, obtains corresponding sampling unit; Wherein, described sampling unit comprises the pixel of even number line even column or even number line odd column or odd-numbered line odd column or odd-numbered line even column in described coding unit.
In a kind of application example of the present invention, for the original pixels block to be encoded of a 2N × 2N, the each one times of down-sampling of horizontal and vertical can be carried out to this original pixels block, namely the pixel of each even number line even column coordinate is retained, and abandon rest of pixels, the sampled pixel block (namely without the input pixel of the current pixel block or image that damage compression) of such generation N × N, i.e. corresponding sampling unit of the present invention.
Certainly, in concrete enforcement, when each one times of down-sampling of horizontal and vertical is carried out to coding unit, the pixel of even number line odd column or odd-numbered line odd column or odd-numbered line even column can also be retained to generate corresponding sampling unit, the present invention does not do concrete restriction to this, in addition, carrying out one times of down-sampling is that the down-sampling the present invention specifically carrying out several times does not limit in order to ensure the quality of picture reducing data volume while.
As can be seen from above-mentioned down-sampling process, the coding unit of a 2N × 2N is obtained to the sampling unit of a N × N through down-sampling, pixel to be encoded has reduced to 1/4th of original pixels, therefore, for predictive mode selection algorithm, need the data volume of computing to decrease 3/4ths, thus the complexity of computing can be reduced.In addition, because the distance spatially between pixel is nearer, their correlation is also stronger, so, there is very strong correlation, Downsapling method of the present invention between neighbor, retain respectively in horizontal and vertical direction or abandon adjacent pixel, under guarantee does not affect the prerequisite of image quality, the complexity of prediction can be reduced.
Step 102, utilize in multiple predictive mode respectively each described sampling unit is predicted, obtain corresponding predicting unit;
The system of selection of traditional intra prediction mode is for each in multiple predictive mode, coding unit is encoded, then calculate the rate distortion costs that each predictive mode is corresponding, and the minimum predictive mode of selection rate distortion cost is optimum prediction mode.But coding is very complicated with rate distortion costs computational process, and in order to accelerate the speed of model selection, the present invention uses pre-selected pattern algorithm.Particularly, can for each in multiple predictive mode, sampling unit is generated to the predicting unit of its correspondence, then the prediction cost between each predicting unit and coding unit is calculated, one or more selecting prediction Least-cost as preliminary election predictive mode calculation rate distortion cost, finally can determine rate distortion costs minimum for optimum prediction mode.By using pre-selected pattern algorithm, avoid and all need first to encode again according to all predictive mode calculation rate distortion costs to all coding units, therefore greatly reduce the number of times of coding and calculation rate distortion cost, thus the complexity of predictive mode selection algorithm can be reduced to a great extent.
In a kind of application example of the present invention, multiple predictive mode can comprise 35 kinds, wherein, specifically can comprise 33 kinds of direction modes (Angular pattern), a kind of DC mode (DC pattern) and a kind of plane mode (Planar pattern).
For a certain pattern x in 35 kinds of predictive modes, can predict for each sampling unit, obtain the predicting unit of corresponding N × N, finally can generate the predicting unit of 35 N × N.Prediction samples computational methods during the same 2N × 2N of its Forecasting Methodology, but the N × N sampling only retaining corresponding down-sampling, therefore computation complexity decreases about 3/4ths.
See Fig. 2, show a kind of schematic diagram utilizing 35 kinds of predictive modes to predict block of pixels of the present invention.
See Fig. 2 a, current block of pixels to be encoded is middle square block, and the gray pixels of upper left is reference pixel, and they are by coding reconstruct, and the prediction of current block produces according to these reference pixels.For any one in 33 kinds of directional prediction modes, first a direction is described, then a line every in current block or each row is all obtained according to the grey reference pixel coding of this direction according to upper left.As Fig. 2 a, for every a line, be mapped to top reference line according to direction, then obtain according to top reference line pixel coder, if point to sub-pixel position, then obtain by bilinear interpolation.Fig. 2 b shows row predictive mode, and each row in current block are mapped to coding on left column according to the direction that illustrates in advance and obtain.If reference columns pixel is pointed to outside left column pixel, so its pixel value obtains from top line mapping according to this direction, as shown in Figure 2 c.Fig. 2 d shows all 33 directional prediction modes.DC pattern is averaged upper left reference pixel exactly, and this mean value is exactly the DC prediction of current block.Planar pattern is exactly generate a horizontal gradient according to upper left reference pixel one to make peace the consistent line by line prediction of vertical gradient by column.
Step 103, calculate prediction cost between each predicting unit and described coding unit;
In one preferred embodiment of the invention, the step of the prediction cost between each predicting unit of described calculating and described coding unit, specifically can comprise:
S1, calculate Adama cost parameter between each predicting unit and described coding unit;
Wherein, the computational process of Adama cost parameter specifically can be as follows:
For a block of pixels of arbitrary size, if it wide and high is all the multiple of 8, so it is divided into some 8 × 8 sub-blocks, otherwise is divided into some 4 × 4 sub-blocks.Then Adama (hadamard) conversion is performed to each 8 × 8 sub-blocks or 4 × 4 sub-blocks, again the absolute value of all conversion coefficients is added up and obtain Ap, if current block is 4 × 4, so (Ap+1)/2 are exactly the hadamard cost of 4 × 4 pieces, if 8 × 8 pieces, so (Ap+2)/4 be exactly current 8 × 8 pieces hadamard cost, a block hadamard cost be its all 4 × 4 or 8 × 8 sub-blocks be partitioned into hadamard cost and.
S2, the prediction cost spending each predicting unit of calculating corresponding according to the Adama cost after weighting and code check.
Particularly, prediction cost corresponding to each predicting unit can be gone out according to the above-mentioned Adama cost parameter calculated according to formulae discovery below:
cost(x)＝(HAD(x)×12□3)+λ
motion×bit(x) (1)
Wherein, HAD (x) is under a certain pattern x in 35 kinds of predictive modes, the Adama cost parameter between each predicting unit and described coding unit, therefore, and total total 35 corresponding prediction costs, λ
motionfor LaGrange parameter,
it is fixing, bit (x) for current prediction mode x corresponding code check cost.
The embodiment of the present invention is in the process calculating prediction cost corresponding to each predicting unit, ranking operation has been carried out to HAD (x), i.e. HAD (x) × 12 3, also namely the value of Adama cost parameter is multiplied by 12/8ths, the object of described ranking operation is the Adama cost parameter in order to simulate 2N × 2N, when not adopting down-sampling algorithm described in the invention is the Adama cost parameter of 2N × 2N, the Adama cost parameter of N × N now, but due to the span of Hadamard transform larger (perform on one of them 8x8 piecemeal of N × N Hadamard transform be equivalent to up-sampling after Hadamard transform on 16x16 block on 2N × 2N block), correlation is lower so the span value of Hadamard transform is larger, so what adopt in the present invention is not take advantage of 4, but only take advantage of 12/8ths.Therefore, the embodiment of the present invention computes weighted to described Adama cost parameter, make result of calculation more accurate, thus can reduce and down-sampling is carried out for unit to be encoded and the loss brought, when can ensure that the overall complexity of encoding reduces about 15% further, and loss of compression efficiency is only below 0.1%.
Step 104, according to prediction cost corresponding to each predicting unit, from described multiple predictive mode, select preliminary election predictive mode;
In one preferred embodiment of the invention, the described prediction cost corresponding according to each predicting unit, from described multiple predictive mode, select preliminary election predictive mode, comprising:
According to the prediction cost that each predicting unit is corresponding, from described multiple predictive mode select prediction Least-cost one or more as preliminary election predictive mode.
Particularly, from 35 predictive modes, select minimum one or more of cost as preliminary election predictive mode.Several preliminary election of concrete selection predictive mode is relevant with the size of current coded unit, and in general, larger block tends to the preliminary election predictive mode adopting number less, and less block then tends to adopt the more preliminary election predictive mode of number.Such as, the coding unit for 4 × 4 or 8 × 8 can select 5 kinds, can select 3 kinds for 16 × 16 and above coding unit.Concrete selection is several can be configured according to actual conditions, the fast algorithm in original HEVC test model be 4 × 4 or 8 × 8 coding unit can select 8 kinds, but do not need so many in reality, therefore, the present invention has made 5 kinds into.
Step 105, calculate corresponding rate distortion costs for described preliminary election predictive mode, and from all preliminary election predictive modes, select best predictive mode according to described rate distortion costs.
Particularly, corresponding rate distortion costs can be calculated for 3 kinds that select in abovementioned steps or 5 kinds of preliminary election predictive modes, and the predictive mode as the best that selection rate distortion cost is minimum from all preliminary election predictive modes.
Wherein, rate distortion costs can calculate according to following formula:
J
mode(x)＝SSD(x)+λ
mode×bit(x) (2)
Wherein, J
modex () represents that the rate distortion costs that preliminary election predictive mode x is corresponding, bit (x) are the bit rates used required for use preliminary election predictive mode x encodes to current coded unit, λ
modebe Lagrange factor, specifically can pass through following formulae discovery:
Wherein, W
kdepend on coding configuration schemes and present image at GOP (Group Of Picture, the length of image sets) in the QP (QuantizationParameter of encoded hierarchy, quantization parameter) side-play amount, its value is between 0.4 to 2, the value of α depends on whether current encoded image is reference picture and B frame number, the value of α can between 0.5 to 1, if current encoded image is non-reference picture, then α=1, otherwise the value of α can pass through following formulae discovery:
α＝1-clip3(0，0.5，0.05×num
bfrm) (4)
Wherein, num
bfrmbe B frame number in a GOP, such as, when encoded hierarchy is 8, num
bfrmdifference of two squares sum between the pixel in original pixels block that to be 7, SSD be and the reconstructed pixel after coding.
To sum up, embodiment of the present invention tool has the following advantages:
First, the embodiment of the present invention carries out down-sampling to coding unit, obtain corresponding sampling unit, and the selection of predictive mode is carried out for sampling unit, pixel due to sampling unit is less than the pixel of coding unit, therefore the calculative data volume of selection algorithm can be simplified, thus the complexity of selection algorithm can be reduced;
In addition, the embodiment of the present invention is not directly for all predictive mode calculation rate distortion costs, but first from described multiple predictive mode, selects preliminary election predictive mode, then for described preliminary election predictive mode calculation rate distortion cost; This, relative to for all predictive mode calculation rate distortion costs, greatly reduces the data volume involved by calculation rate distortion cost, thus can reduce the complexity of selection algorithm further;
Further, owing to having very strong correlation between neighbor, the present invention carries out each one times of down-sampling respectively in horizontal and vertical direction, under guarantee does not affect the prerequisite of image quality, can reduce the complexity of prediction;
Finally, the present invention is when the prediction cost that each predicting unit of calculating is corresponding, primal algorithm is improved, namely prediction cost corresponding to each predicting unit is calculated by computing weighted to described Adama cost parameter, make result of calculation more accurate, thus can reduce and down-sampling is carried out for unit to be encoded and the loss that brings, when can ensure that the overall complexity of encoding reduces about 15% further, and loss of compression efficiency is only below 0.1%.
Device embodiment
With reference to Fig. 3, show the structured flowchart of the choice device embodiment of a kind of intra prediction mode of the present invention, specifically can comprise:
Sampling module 310, for carrying out down-sampling to coding unit, obtains corresponding sampling unit;
Prediction module 320, predicts described sampling unit for each utilizing in multiple predictive mode respectively, obtains corresponding predicting unit;
Computing module 330, for calculating the prediction cost between each predicting unit and described coding unit;
Preliminary election module 340, for according to prediction cost corresponding to each predicting unit, selects preliminary election predictive mode from described multiple predictive mode; And
Selecting module 350, for calculating corresponding rate distortion costs for described preliminary election predictive mode, and from all preliminary election predictive modes, selecting best predictive mode according to described rate distortion costs.
In one preferred embodiment of the invention, described sampling module 310, comprising:
Sampling unit, for carrying out each one times of down-sampling of horizontal and vertical respectively to coding unit, obtains corresponding sampling unit; Wherein, described sampling unit comprises the pixel of even number line even column or even number line odd column or odd-numbered line odd column or odd-numbered line even column in described coding unit.
In another preferred embodiment of the invention, described computing module 330, comprising:
First computing unit, for calculate between each predicting unit and described coding unit Adama cost parameter;
Second computing unit, for calculating prediction cost corresponding to each predicting unit according to the Adama cost after weighting and code check cost.
In another preferred embodiment of the present invention, described preliminary election module 340, comprising:
Pre-selection unit, for according to prediction cost corresponding to each predicting unit, one or more selecting prediction Least-cost from described multiple predictive mode are as preliminary election predictive mode.
In another preferred embodiment of the present invention, described selection module 350, comprising:
Cost calculating unit, for calculating corresponding rate distortion costs for described preliminary election predictive mode;
Pattern determining unit, for determining that the minimum preliminary election predictive mode of rate distortion costs is best predictive mode.
For device embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.
Each embodiment in this specification all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.
Those skilled in the art should understand, the embodiment of the embodiment of the present invention can be provided as method, device or computer program.Therefore, the embodiment of the present invention can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.And the embodiment of the present invention can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disc store, CD-ROM, optical memory etc.) of computer usable program code.
The embodiment of the present invention describes with reference to according to the flow chart of the method for the embodiment of the present invention, terminal equipment (system) and computer program and/or block diagram.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block diagram and/or square frame and flow chart and/or block diagram and/or square frame.These computer program instructions can being provided to the processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing terminal equipment to produce a machine, making the instruction performed by the processor of computer or other programmable data processing terminal equipment produce device for realizing the function of specifying in flow chart flow process or multiple flow process and/or block diagram square frame or multiple square frame.
These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing terminal equipment, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in flow chart flow process or multiple flow process and/or block diagram square frame or multiple square frame.
These computer program instructions also can be loaded on computer or other programmable data processing terminal equipment, make to perform sequence of operations step to produce computer implemented process on computer or other programmable terminal equipment, thus the instruction performed on computer or other programmable terminal equipment is provided for the step realizing the function of specifying in flow chart flow process or multiple flow process and/or block diagram square frame or multiple square frame.
Although described the preferred embodiment of the embodiment of the present invention, those skilled in the art once obtain the basic creative concept of cicada, then can make other change and amendment to these embodiments.So claims are intended to be interpreted as comprising preferred embodiment and falling into all changes and the amendment of embodiment of the present invention scope.
Finally, also it should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operating space, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or terminal equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or terminal equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the terminal equipment comprising described key element and also there is other identical element.
Above to system of selection and the device of a kind of intra prediction mode provided by the present invention, be described in detail, apply specific case herein to set forth principle of the present invention and execution mode, the explanation of above embodiment just understands method of the present invention and core concept thereof for helping; Meanwhile, for one of ordinary skill in the art, according to thought of the present invention, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (10)
1. a system of selection for intra prediction mode, is characterized in that, described method comprises:
Down-sampling is carried out to coding unit, obtains corresponding sampling unit;
Each utilizing in multiple predictive mode is respectively predicted described sampling unit, obtains corresponding predicting unit;
Calculate the prediction cost between each predicting unit and described coding unit;
According to the prediction cost that each predicting unit is corresponding, from described multiple predictive mode, select preliminary election predictive mode;
Calculate corresponding rate distortion costs for described preliminary election predictive mode, and from all preliminary election predictive modes, select best predictive mode according to described rate distortion costs.
2. the method for claim 1, is characterized in that, describedly carries out down-sampling to coding unit, obtains corresponding sampling unit, comprising:
The each one times of down-sampling of horizontal and vertical is carried out respectively to coding unit, obtains corresponding sampling unit; Wherein, described sampling unit comprises the pixel of even number line even column or even number line odd column or odd-numbered line odd column or odd-numbered line even column in described coding unit.
3. the method for claim 1, is characterized in that, the prediction cost between each predicting unit of described calculating and described coding unit, comprising:
Calculate between each predicting unit and described coding unit Adama cost parameter;
Prediction cost corresponding to each predicting unit is calculated according to the Adama cost after weighting and code check cost.
4. the method for claim 1, is characterized in that, the described prediction cost corresponding according to each predicting unit, selects preliminary election predictive mode, comprising from described multiple predictive mode:
According to the prediction cost that each predicting unit is corresponding, from described multiple predictive mode select prediction Least-cost one or more as preliminary election predictive mode.
5. the method for claim 1, is characterized in that, the described rate distortion costs calculating correspondence for described preliminary election predictive mode, and from all preliminary election predictive modes, selects best predictive mode according to described rate distortion costs, comprising:
Corresponding rate distortion costs is calculated for described preliminary election predictive mode;
Determine that the minimum preliminary election predictive mode of rate distortion costs is best predictive mode.
6. a choice device for intra prediction mode, is characterized in that, described device comprises:
Sampling module, for carrying out down-sampling to coding unit, obtains corresponding sampling unit;
Prediction module, predicts described sampling unit for each utilizing in multiple predictive mode respectively, obtains corresponding predicting unit;
Computing module, for calculating the prediction cost between each predicting unit and described coding unit;
Preliminary election module, for according to prediction cost corresponding to each predicting unit, selects preliminary election predictive mode from described multiple predictive mode; And
Selecting module, for calculating corresponding rate distortion costs for described preliminary election predictive mode, and from all preliminary election predictive modes, selecting best predictive mode according to described rate distortion costs.
7. device as claimed in claim 6, it is characterized in that, described sampling module, comprising:
Sampling unit, for carrying out each one times of down-sampling of horizontal and vertical respectively to coding unit, obtains corresponding sampling unit; Wherein, described sampling unit comprises the pixel of even number line even column or even number line odd column or odd-numbered line odd column or odd-numbered line even column in described coding unit.
8. device as claimed in claim 6, it is characterized in that, described computing module, comprising:
First computing unit, for calculate between each predicting unit and described coding unit Adama cost parameter;
Second computing unit, for calculating prediction cost corresponding to each predicting unit according to the Adama cost after weighting and code check cost.
9. device as claimed in claim 6, it is characterized in that, described preliminary election module, comprising:
Pre-selection unit, for according to prediction cost corresponding to each predicting unit, one or more selecting prediction Least-cost from described multiple predictive mode are as preliminary election predictive mode.
10. device as claimed in claim 6, it is characterized in that, described selection module, comprising:
Cost calculating unit, for calculating corresponding rate distortion costs for described preliminary election predictive mode;
Pattern determining unit, for determining that the minimum preliminary election predictive mode of rate distortion costs is best predictive mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410850621.9A CN104754338A (en) | 2014-12-30 | 2014-12-30 | Selection method and device for intra-frame predication mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410850621.9A CN104754338A (en) | 2014-12-30 | 2014-12-30 | Selection method and device for intra-frame predication mode |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104754338A true CN104754338A (en) | 2015-07-01 |
Family
ID=53593359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410850621.9A Pending CN104754338A (en) | 2014-12-30 | 2014-12-30 | Selection method and device for intra-frame predication mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104754338A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208391A (en) * | 2015-09-17 | 2015-12-30 | 广州市百果园网络科技有限公司 | Intra-frame prediction method and device |
| CN105681812A (en) * | 2016-03-30 | 2016-06-15 | 腾讯科技（深圳）有限公司 | HEVC (high efficiency video coding) intra-frame coding processing method and device |
| CN107426573A (en) * | 2017-08-08 | 2017-12-01 | 鄂尔多斯应用技术学院 | Adaptive fast prediction dividing elements method and device based on motion homogeney |
| CN107613294A (en) * | 2017-10-12 | 2018-01-19 | 杭州当虹科技有限公司 | A kind of method for fast skipping P, B frame intra prediction mode in HEVC |
| CN108156458A (en) * | 2017-12-28 | 2018-06-12 | 北京奇艺世纪科技有限公司 | A kind of determining coding mode method and device |
| WO2018233411A1 (en) * | 2017-06-23 | 2018-12-27 | 腾讯科技（深圳）有限公司 | Prediction mode selection method, video encoding device and storage medium |
| CN109495742A (en) * | 2018-12-13 | 2019-03-19 | 深圳市网心科技有限公司 | A kind of video frame coding method, device and equipment |
| CN109672895A (en) * | 2018-12-27 | 2019-04-23 | 北京佳讯飞鸿电气股份有限公司 | A kind of HEVC intra-frame prediction method and system |
| WO2019119572A1 (en) * | 2017-12-22 | 2019-06-27 | 深圳市华星光电半导体显示技术有限公司 | Compensation table compression method |
| CN114885164A (en) * | 2022-07-12 | 2022-08-09 | 深圳比特微电子科技有限公司 | Method and device for determining intra-frame prediction mode, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102196272A (en) * | 2010-03-11 | 2011-09-21 | 中国科学院微电子研究所 | P frame encoding method and device |
| CN103997646A (en) * | 2014-05-13 | 2014-08-20 | 北京航空航天大学 | Rapid intra-frame prediction mode selection method in high-definition video coding |
-
2014
- 2014-12-30 CN CN201410850621.9A patent/CN104754338A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102196272A (en) * | 2010-03-11 | 2011-09-21 | 中国科学院微电子研究所 | P frame encoding method and device |
| CN103997646A (en) * | 2014-05-13 | 2014-08-20 | 北京航空航天大学 | Rapid intra-frame prediction mode selection method in high-definition video coding |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208391A (en) * | 2015-09-17 | 2015-12-30 | 广州市百果园网络科技有限公司 | Intra-frame prediction method and device |
| CN105681812B (en) * | 2016-03-30 | 2019-11-19 | 腾讯科技（深圳）有限公司 | HEVC intraframe coding treating method and apparatus |
| CN105681812A (en) * | 2016-03-30 | 2016-06-15 | 腾讯科技（深圳）有限公司 | HEVC (high efficiency video coding) intra-frame coding processing method and device |
| WO2018233411A1 (en) * | 2017-06-23 | 2018-12-27 | 腾讯科技（深圳）有限公司 | Prediction mode selection method, video encoding device and storage medium |
| CN109120926A (en) * | 2017-06-23 | 2019-01-01 | 腾讯科技（深圳）有限公司 | Predicting mode selecting method, device and video encoder |
| US10986332B2 (en) * | 2017-06-23 | 2021-04-20 | Tencent Technology (Shenzhen) Company Limited | Prediction mode selection method, video encoding device, and storage medium |
| US20190230349A1 (en) * | 2017-06-23 | 2019-07-25 | Tencent Technology (Shenzhen) Company Limited | Prediction mode selection method, video encoding device, and storage medium |
| CN109120926B (en) * | 2017-06-23 | 2019-08-13 | 腾讯科技（深圳）有限公司 | Predicting mode selecting method, device and medium |
| CN107426573A (en) * | 2017-08-08 | 2017-12-01 | 鄂尔多斯应用技术学院 | Adaptive fast prediction dividing elements method and device based on motion homogeney |
| CN107613294A (en) * | 2017-10-12 | 2018-01-19 | 杭州当虹科技有限公司 | A kind of method for fast skipping P, B frame intra prediction mode in HEVC |
| CN107613294B (en) * | 2017-10-12 | 2020-04-07 | 杭州当虹科技股份有限公司 | Method for rapidly skipping P, B frame intra-frame prediction mode in HEVC |
| WO2019119572A1 (en) * | 2017-12-22 | 2019-06-27 | 深圳市华星光电半导体显示技术有限公司 | Compensation table compression method |
| CN108156458A (en) * | 2017-12-28 | 2018-06-12 | 北京奇艺世纪科技有限公司 | A kind of determining coding mode method and device |
| CN108156458B (en) * | 2017-12-28 | 2020-04-10 | 北京奇艺世纪科技有限公司 | Method and device for determining coding mode |
| CN109495742A (en) * | 2018-12-13 | 2019-03-19 | 深圳市网心科技有限公司 | A kind of video frame coding method, device and equipment |
| CN109672895A (en) * | 2018-12-27 | 2019-04-23 | 北京佳讯飞鸿电气股份有限公司 | A kind of HEVC intra-frame prediction method and system |
| CN114885164A (en) * | 2022-07-12 | 2022-08-09 | 深圳比特微电子科技有限公司 | Method and device for determining intra-frame prediction mode, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104754338A (en) | Selection method and device for intra-frame predication mode | |
| KR102073638B1 (en) | Picture prediction method and picture prediction device | |
| KR101682148B1 (en) | Method and apparatus for interpolation based on transform and inverse transform | |
| US9451266B2 (en) | Optimal intra prediction in block-based video coding to calculate minimal activity direction based on texture gradient distribution | |
| CN110557631B (en) | Image prediction method and related equipment | |
| AU2015213341B2 (en) | Video decoder, video encoder, video decoding method, and video encoding method | |
| WO2011064990A1 (en) | Image encoding apparatus, image decoding apparatus, image encoding method, and image decoding method | |
| CN112055203B (en) | Inter-frame prediction method, video coding method and related devices | |
| TW201635797A (en) | Method, apparatus and coder for selecting optimal reference frame in hevc coding | |
| CN108989799B (en) | Method and device for selecting reference frame of coding unit and electronic equipment | |
| CN111263144B (en) | Motion information determination method and equipment | |
| CN103702122A (en) | Coding mode selection method, device and coder | |
| CN103596003B (en) | Interframe predication quick mode selecting method for high-performance video coding | |
| MXPA05007303A (en) | Method and device for choosing a mode of coding. | |
| CN104918047B (en) | A kind of method and device for removing of bi-directional motion estimation | |
| CN102801982A (en) | Estimation method applied on video compression and based on quick movement of block integration | |
| JP7439219B2 (en) | Intra prediction method and device, computer readable storage medium | |
| CN104601991A (en) | Searching method and device for time-domain predictive parameters | |
| US10187656B2 (en) | Image processing device for adjusting computational complexity of interpolation filter, image interpolation method, and image encoding method | |
| CN105122808A (en) | Method and apparatus of disparity vector derivation for three-dimensional and multi-view video coding | |
| KR20040035390A (en) | Method and apparatus for motion estimation using of adaptive search pattern for video sequence compression | |
| TW201501511A (en) | Prediction method and system in image compression | |
| CN109600608B (en) | Dual-mode selection prediction method for complex texture in bandwidth compression | |
| CN105306953A (en) | Image coding method and device | |
| JP2007243859A (en) | Image encoding apparatus and image encoding program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150701 |