CN100518324C - Conversion method from compression domain MPEG-2 based on interest area to H.264 video - Google Patents
Conversion method from compression domain MPEG-2 based on interest area to H.264 video Download PDFInfo
- Publication number
- CN100518324C CN100518324C CN 200710044220 CN200710044220A CN100518324C CN 100518324 C CN100518324 C CN 100518324C CN 200710044220 CN200710044220 CN 200710044220 CN 200710044220 A CN200710044220 A CN 200710044220A CN 100518324 C CN100518324 C CN 100518324C
- Authority
- CN
- China
- Prior art keywords
- interest
- area
- compression domain
- conversion
- coefficient
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The method comprises: making the conversion for the intra-frame coding frame; only making the transformation from the compress domain digital cosine transform coefficient to the integer transform coefficient; making transformation for the compress domain intra-frame prediction. Making detection for the interested area in the compress domain: the summation of absolute values of the horizontal component and vertical component of current macro block motion vector is less than or equals the threshold TH, then the macro block is estimated as an interested area. Making transformation for the non-interested area: making compress domain open-loop transformation; by using the transformation from the intra compress domain digital cosine transform coefficient to the integer transform coefficient, making re-quantization. Making transformation for the interested area: in the compress domain, making quantization error compensation; making the interpolation error compensation close-loop transformation.
Description
Technical field
What the present invention relates to is the method for the video conversion in a kind of telecommunication technology field, more specifically, relates to a kind of the realization and arrives the H.264 method of video conversion based on the compression domain MPEG-2 of area-of-interest.
Background technology
The video switch technology is that a kind of compressed format video flowing is converted into another kind of compressed format.At present,, exist the various video coding standard towards various field of video applications, such as the Moving Picture Experts Group-2 that Motion Picture Experts Group's (MPEG tissue) formulates, MPEG-4 standard and standard etc. H.264.How can be efficient, the realization conversion of high-quality makes it to be adapted in the different hardware equipment, is the problem that industry is paid close attention to most always.H.264 the standard of uniting joint video team (JVT tissue) formulation of establishment by Motion Picture Experts Group's (MPEG tissue) and video coding expert group (VCEG tissue), as video encoding standard of new generation, improvement on many performances is provided, will have progressively occupied the leading position of field of video applications.Yet present existing a large amount of MPEG-2 video program sources, and the encoder and the set-top box device of millions of MPEG-2 forms.Therefore, MPEG-2 and H.264 two kinds of video standards will in for a long time, coexist, MPEG-2 is converted into the current research focus that presses for solution to video H.264.
Find through literature search prior art, people such as Tuanjie Qian are at " IEEE video technology circuit and system's transactions " (IEEE Transactions on Circuits and Systems for VideoTechnology), Vol.16, NO.4, pp.523-534, among " compression domain MPEG-2 of interpolation error compensation is to H.264 conversion " (Transform Domain Transcoding FromMPEG-2to is the With Interpolation Drift-Error Compensation H.264) that delivers on the April.2006, the compression domain MPEG-2 that proposes the interpolation error compensation arrives H.264 video switch technology, improved the video quality of compression domain conversion greatly, its core concept is on the error theory analysis foundation to the compression domain conversion, outside to quantization error compensation, at special interpolation algorithm H.264, proposed to carry out the method for the compression domain conversion of interpolation error compensation.But owing to need do quantization error and interpolation error compensation in compression domain, the computational complexity of this method also increases greatly, and just improper to the requirement of real-time conversion, this also is the drawback of this method.
Summary of the invention
The present invention is directed to the deficiencies in the prior art, proposed a kind of the realization and arrived the H.264 method of video conversion based on the compression domain MPEG-2 of area-of-interest, combine the advantage of compression domain open loop and closed-loop policy, make it can realize real-time conversion, saved the speed of compression domain conversion greatly, and the well as subjective video quality of conversion does not reduce.
The present invention is achieved by the following technical solutions, and concrete steps are as follows:
The first step, the conversion of intracoded frame (I frame) compression domain
Intracoded frame does not have motion vector, can only utilize the compression domain coefficient to carry out method such as border estimation carries out area-of-interest and detects, but because intracoded frame need not passed through inter motion compensation, therefore do not need to carry out the compensation of compression domain interpolation error, directly carry out the compression domain infra-frame prediction, just can obtain converting video quality preferably at H.264 special infra-frame prediction.Therefore, in the inventive method, not carrying out area-of-interest at intracoded frame detects, only carry out common compression domain weightization and the compression domain infra-frame prediction is changed, promptly only carry out common compression domain number cosine converting (DCT) coefficient and change, just can obtain converting video quality preferably to integer transform (IT), weightization and compression domain infra-frame prediction.
In second step, the compression domain area-of-interest detects
The notion of area-of-interest is to putting forward on the basis of human visual system (HVS) research.Discover the degree of concern difference of human eye for entire image, also far from each other to the resolving power of zones of different.Characteristic according to the human visual system, human eye more interesting areas is mainly determined by movable information, spatial frequency information, boundary information and colouring information, for video sequence, movable information detects particularly important to area-of-interest, can think that area-of-interest is made up of moving region or target.And conversion is at the video code flow after the compression, except that intracoded frame, can utilize the motion vector information in the code stream easily, is not increasing substantially under the situation of translation operation amount, realizes the detection quick and precisely of compression domain area-of-interest.
Therefore, among the present invention, the motion vector MVs that utilizes the MPEG-2 code stream decoding to obtain judges whether to be area-of-interest one by one to each macro block in the code stream.The zone that does not move just is judged as the background area, i.e. the insensitive zone of human eye vision; The moving region just is judged as visual attention concentrated area, i.e. area-of-interest.Concrete Rule of judgment is: if the motion vector of current macro (i piece) | MVx (i) |+| MVy (i) | smaller or equal to threshold value TH, then current macro is non-area-of-interest, otherwise is area-of-interest.Determine after the ROI, just can adopt the different switching method, obtain the better compromise of operand and video quality at zones of different.
The 3rd step, the conversion of non-area-of-interest
The open loop conversion is exactly without motion compensation in compression domain, directly passes through the interior DCT coefficient of compression domain to IT transformation of coefficient, re-quantization again.Can save a large amount of computational complexities like this, cost is not carry out error compensation, can cause tangible error drift, makes video quality descend.Yet as noted earlier, large-area non-area-of-interest human eye and insensitive reduces the decline that this regional video quality can not cause tangible whole video subjective quality in the video sequence.
Therefore, the inventive method adopts compression domain open loop conversion to non-area-of-interest, make in this way, more macroblock skip a large amount of matrix multiplication processes of compression domain quantization error compensation and the compensation of compression domain interpolation error, saved amount of calculation greatly; And the decline of the video quality in the insensitive zone of human eye can not cause the decline of tangible whole subjective quality.
The 4th step, the conversion of area-of-interest
The video quality that adopts the compression domain motion compensation can avoid error drift to cause during conversion descends, owing to H.264 adopted the interpolation method different with MPEG-2, therefore beyond the quantization error compensation of conventional compression territory, also need to carry out the compensation of compression domain interpolation error, be after the DCT coefficient is converted to the IT coefficient, carry out quantization error compensation in compression domain, and interpolation error compensation, re-quantization again.This process operand is bigger, can obtain video quality preferably, can be suitable with pixel domain converting video quality.
In order to improve the subjective quality of whole converting video, the inventive method compensates the conversion quality that improves area-of-interest by area-of-interest is carried out the compression domain interpolation error, can between conversion quality and operand, keep a balance, guarantee to guarantee that whole subjective quality can not descend in the conversion real-time.
Generally, comprised the larger area zone in the high-resolution video sequence to human eye and insensitive, the quality that improves area-of-interest can be improved the subjective quality of whole video sequence to a great extent.Therefore, among the present invention, adopt different converting structures with non-area-of-interest, the insensitive non-area-of-interest of human eye is carried out compression domain open loop conversion, can accelerate conversion speed greatly, save operand at area-of-interest; Area-of-interest is carried out the conversion of compression domain interpolation compensation closed loop, improve the video conversion quality of area-of-interest, thereby when saving operand, improve the subjective quality of converting video, obtain the better compromise of operand and converting video quality, this is the core concept of the inventive method.
In the methods of the invention, making full use of human visual system's characteristic carries out changing based on the video of area-of-interest.For the insensitive large-area background area of human eye, can adopt quick open loop conversion method, the decline of this zone video quality can not cause the decline of tangible whole video subjective quality.In view of this consideration, when conversion, only need carry out the compensation of compression domain interpolation error, so just can improve the video quality of area-of-interest, take into account the speed and the video quality of conversion the area-of-interest that detects.And for very crucial area-of-interest determine that the present invention has adopted a kind of compression domain detection method of simple and flexible, can when not increasing the conversion complexity, realize that area-of-interest detects fast and accurately.Simulation result shows, the present invention is better than the method for existing MPEG-2 to the conversion of video H.264: whole conversion speed substantially exceeds the conversion speed of existing algorithm, average conversion speed has improved about 5 times than reference image prime field cascade conversion method, improved about 2 times than compression domain interpolation error compensation conversion method, and the video quality of conversion does not reduce.
Description of drawings
Fig. 1 is the inventive method flow chart.
Fig. 2 is compression domain open loop flow path switch figure.
Fig. 3 is compression domain interpolation error compensation closed loop flow path switch figure.
Embodiment
Below in conjunction with accompanying drawing embodiments of the invention are elaborated: present embodiment has provided detailed execution mode and process being to implement under the prerequisite with the technical solution of the present invention, but protection scope of the present invention is not limited to following embodiment.
As shown in Figure 1, present embodiment can be divided into following step:
(1) compression domain of intracoded frame conversion
The MPEG-2 input code flow is carried out the entropy decoding, can conveniently obtain the type of coding of every frame, adopt the different switching mode at the different coding type frame.Concrete grammar is as follows:
The entropy decoding obtains the type of coding of present frame.
If present frame is an intracoded frame, then not carrying out area-of-interest detects, only carry out the MPEG-2 inverse quantization and obtain the DCT coefficient, carry out DCT then to the IT transformation of coefficient, obtain the IT coefficient, the IT coefficient is carried out the compression domain infra-frame prediction again, H.264 the IT coefficient after the prediction again through quantizing and entropy coding, generate H.264 code stream, just can obtain conversion quality preferably.
If present frame is an encoded predicted frame, comprise forward predicted frame and bi-directional predicted frames, then carry out the compression domain area-of-interest earlier and detect, adopt the different switching mode at zones of different again.
Because intracoded frame does not adopt motion compensation, need not to adopt interpolation error to compensate and avoid error drift, therefore, use the present embodiment method, not needing to carry out area-of-interest during the intracoded frame conversion detects, directly adopt weightization and compression domain infra-frame prediction, just can obtain converting video quality preferably.
(2) detection of compression domain area-of-interest
The MPEG-2 input code flow is carried out the entropy decoding, can obtain each macroblock encoding pattern and motion vector MV.Motion vector MV with current macro judges that whether this macro block is area-of-interest, judges one by one to each macro block in the present frame.Concrete grammar is as follows:
The motion vector MV (i) of calculating current macro=| MVx (i) |+| MVy (i) |
Wherein i represents i macro block in the present frame,
The horizontal component of i macro block motion vector of MVx (i) expression,
The vertical component of i macro block motion vector of MVy (i) expression.
If MV (i) is smaller or equal to threshold value TH, then
Current i macro block is judged as non-area-of-interest.
If MV (i) is greater than threshold value TH, then
Current i macro block is judged as area-of-interest.
After the area-of-interest of having determined every frame, just can adopt the different switching framework at zones of different.Use the present embodiment method, area-of-interest detects and realizes in compression domain, fast effectively, and adopt the operand that can't obviously increase conversion after the present embodiment method, for most of cycle tests, the whole relatively translation operation amount of operand that the compression domain area-of-interest detects can be ignored.
(3) conversion of non-area-of-interest
To being judged as the macro block of non-area-of-interest, change over switch during conversion in the accompanying drawing 1 is got to the below heavy line, do not carry out the compensation of compression domain quantization error and interpolation error, after the MPEG-2 inverse quantization obtains the DCT coefficient, process DCT coefficient is to the IT transformation of coefficient, directly carry out H.264 weightization, and entropy coding, as shown in Figure 2.
(4) conversion of area-of-interest
To being judged as the macro block of area-of-interest, the change over switch during conversion in the accompanying drawing 1 is got to the top fine line, carries out the compensation of compression domain quantization error and interpolation error, so that improve this regional converting video quality, thereby improves overall video quality.After the MPEG-2 inverse quantization obtains the DCT coefficient, to the IT transformation of coefficient, carry out the compensation of compression domain quantization error compensation and interpolation error through the DCT coefficient respectively, carry out H.264 weightization again, and entropy coding, as shown in Figure 3.Wherein when compression domain quantization error compensation and interpolation error compensation, the predicted macroblock of motion vectors point has three kinds of situations:
● if predicted macroblock is the area-of-interest macro block, then directly carries out compression domain error compensation;
● if predicted macroblock partly is an area-of-interest, and part is non-area-of-interest, then earlier non-area-of-interest part macro block is filled up computing, duplicates the data of adjacent area-of-interest part macro block when filling up, and then carries out compression domain error compensation;
● if predicted macroblock is non-area-of-interest macro block, and then this predicted macroblock is directly duplicated the data of the most adjacent area-of-interest macro block, and then carries out compression domain error compensation.
In the present embodiment, for saving operand, the insensitive non-area-of-interest of human eye is carried out compression domain open loop conversion, because error drift can cause the conversion quality of non-area-of-interest to descend, if expectation further improves the conversion quality of non-area-of-interest, save operand simultaneously as far as possible, can only carry out the compression domain quantization error compensation to this zone, save the operand of compression domain interpolation error compensation, but the speed of whole conversion will decrease; If expectation further improves the conversion quality of area-of-interest, can also when weight H.264, adjust quantization parameter, reduce the quantization parameter of area-of-interest, carefully quantize, improve the video quality of area-of-interest, increase the quantization parameter of non-area-of-interest simultaneously, slightly quantize, reduce the quality of non-area-of-interest, adopt sign indicating number control, can further improve whole converting video quality based on area-of-interest.
Use a plurality of cycle testss of 300 frames that compression domain open loop conversion, compression domain interpolation error compensation closed loop conversion and these three kinds of compression domain MPEG-2s of present embodiment method are carried out emulation to the method for H.264 video conversion, can draw from the statistics of emulation: the present embodiment method is to stationary background, the sequence effect of target travel is all fine, its conversion speed substantially exceeds the conversion of compression domain interpolation error compensation closed loop, and its converting video quality obviously improves than compression domain open loop conversion.If this method is used in combination with sign indicating number control based on area-of-interest, can also when not increasing treating capacity, further improve the converting video quality.
In sum, the present invention is better than other method, and its advantage is: (1) carries out area-of-interest in compression domain Detect, fast effectively, fine to the effect of motion sequence; (2) avoided above-mentioned existing methodical deficiency; (3) Utilize more fully human visual system's characteristic, utilized human eye to large-area non-area-of-interest Quality Down is insensitive, has saved the compression domain error compensation calculating process of non-area-of-interest, can be greatly Improve conversion speed; (4) take full advantage of human eye to the characteristic of moving region sensitivity, improved region of interest The conversion quality in territory makes between converting video quality and the conversion speed to have kept balance. This method can Keep greatly improving existing conversion speed on the basis of whole converting video quality.
Claims (5)
1, a kind of realization arrived the H.264 method of video conversion based on the compression domain MPEG-2 of area-of-interest, it is characterized in that step is as follows:
The first step, the conversion of intracoded frame compression domain
Carry out compression domain number cosine converting coefficient to integer transform transformation of coefficient, compression domain infra-frame prediction, weightization;
In second step, the compression domain area-of-interest detects
Utilize the motion vector information that obtains in the transfer process that each macro block in the frame is judged one by one, the condition that is judged as area-of-interest is: if the motion vector MV (i) of current macro is smaller or equal to threshold value TH, then this macro block is judged as non-area-of-interest, otherwise this macro block is judged as area-of-interest, and wherein the motion vector MV (i) of macro block adopts following formula to weigh:
MV(i)＝|MVx(i)|+|MVy(i)|
MVx in the formula (i) and MVy (i) are respectively the level and the vertical component of i macro block motion vector;
The 3rd step, the conversion of non-area-of-interest
Carry out compression domain open loop conversion,, directly pass through the interior number cosine converting coefficient of compression domain to integer transform transformation of coefficient, re-quantization again without motion compensation;
The 4th step, the conversion of area-of-interest
After the number cosine converting coefficient is converted to the integer transform coefficient, carry out quantization error compensation in compression domain, and interpolation error compensation, re-quantization again.
2, H.264 realization according to claim 1 arrives based on the compression domain MPEG-2 of area-of-interest
The method of video conversion is characterized in that, the described first step, and the conversion of intracoded frame compression domain, specific as follows:
● the entropy decoding obtains the present frame type of coding;
● if present frame is an intracoded frame, then not carrying out area-of-interest detects, only carry out the MPEG-2 inverse quantization and obtain the number cosine converting coefficient, carry out the number cosine converting coefficient then to the integer transform transformation of coefficient, obtain the integer transform coefficient, the integer transform coefficient is carried out the compression domain infra-frame prediction again, and the integer transform coefficient after the prediction through H.264 quantizing and entropy coding, generates H.264 code stream again;
● if present frame is an encoded predicted frame, comprises forward predicted frame and bi-directional predicted frames, then carries out the compression domain area-of-interest earlier and detects, and adopts corresponding conversion regime at the zone again.
3, realization according to claim 1 is arrived the H.264 method of video conversion based on the compression domain MPE6-2 of area-of-interest, it is characterized in that, in described the 3rd step, to being judged as the macro block of non-area-of-interest, do not carry out the compensation of compression domain quantization error and interpolation error during conversion, after the MPEG-2 inverse quantization obtained the number cosine converting coefficient, process number cosine converting coefficient was to the integer transform transformation of coefficient, directly carry out H.264 weightization, and entropy coding.
4, realization according to claim 1 is arrived the H.264 method of video conversion based on the compression domain MPEG-2 of area-of-interest, it is characterized in that, in described the 4th step, to being judged as the macro block of area-of-interest, carry out the compensation of compression domain quantization error and interpolation error during conversion, improve this regional converting video quality, after the MPEG-2 inverse quantization obtains the number cosine converting coefficient, process number cosine converting coefficient is to the integer transform transformation of coefficient, carry out the compensation of compression domain quantization error compensation and interpolation error respectively, carry out H.264 weightization again, and entropy coding.
5, realization according to claim 4 is arrived the H.264 method of video conversion based on the compression domain MPEG-2 of area-of-interest, it is characterized in that, when carrying out the compensation of compression domain quantization error compensation and interpolation error, the predicted macroblock of motion vectors point has three kinds of situations:
● if predicted macroblock is the area-of-interest macro block, then directly carries out compression domain error compensation;
● if predicted macroblock partly is an area-of-interest, and part is non-area-of-interest, then earlier non-area-of-interest part macro block is filled up computing, duplicates the data of adjacent area-of-interest part macro block when filling up, and then carries out compression domain error compensation;
● if predicted macroblock is non-area-of-interest macro block, and then this predicted macroblock is directly duplicated the data of the most adjacent area-of-interest macro block, and then carries out compression domain error compensation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710044220 CN100518324C (en) | 2007-07-26 | 2007-07-26 | Conversion method from compression domain MPEG-2 based on interest area to H.264 video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200710044220 CN100518324C (en) | 2007-07-26 | 2007-07-26 | Conversion method from compression domain MPEG-2 based on interest area to H.264 video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101102492A CN101102492A (en) | 2008-01-09 |
| CN100518324C true CN100518324C (en) | 2009-07-22 |
Family
ID=39036543
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200710044220 Expired - Fee Related CN100518324C (en) | 2007-07-26 | 2007-07-26 | Conversion method from compression domain MPEG-2 based on interest area to H.264 video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100518324C (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101765008B (en) * | 2009-12-28 | 2012-11-14 | 北京工业大学 | Method for encoding and decoding video as well as device and system therefor |
| CN101945281B (en) * | 2010-09-10 | 2014-09-10 | 中兴通讯股份有限公司 | Method and device for filtering video codes |
| CN103002280B (en) * | 2012-10-08 | 2016-09-28 | 中国矿业大学 | Distributed decoding method based on HVS&ROI and system |
| CN104038768B (en) * | 2014-04-30 | 2017-07-18 | 中国科学技术大学 | The many reference field rapid motion estimating methods and system of a kind of coding mode |
| CN104519366B (en) * | 2014-12-04 | 2019-05-14 | 广东中星微电子有限公司 | A kind of Video coding change quantization method and apparatus |
| KR102511705B1 (en) * | 2015-11-16 | 2023-03-20 | 삼성전자주식회사 | Method of encoding video, video encoder performing the same and electronic system including the same |
| CN107454395A (en) * | 2017-08-23 | 2017-12-08 | 上海安威士科技股份有限公司 | A kind of high-definition network camera and intelligent code stream control method |
| CN111491167B (en) * | 2019-10-28 | 2022-08-26 | 华为技术有限公司 | Image encoding method, transcoding method, device, equipment and storage medium |
| CN113453007A (en) * | 2020-03-24 | 2021-09-28 | 合肥君正科技有限公司 | Method for improving monitoring scene H264 coding efficiency |
-
2007
- 2007-07-26 CN CN 200710044220 patent/CN100518324C/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| Transform domain transcoding from MPEG-2 to H.264with interpolation drift-error compensation. Tuanjie Qian, Jun Sun, Dian Li, Xiaokang Yang, Jia Wang.IEEE Transactions on Circuits and Systems for Video Technology,Vol.VOL. 6 No.NO. 4. 2006 |
| Transform domain transcoding from MPEG-2 to H.264with interpolation drift-error compensation. Tuanjie Qian, Jun Sun, Dian Li, Xiaokang Yang, Jia Wang.IEEE Transactions on Circuits and Systems for Video Technology,Vol.VOL. 6 No.NO. 4. 2006 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101102492A (en) | 2008-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100518324C (en) | Conversion method from compression domain MPEG-2 based on interest area to H.264 video | |
| KR100719647B1 (en) | Video data conversion device and video data conversion method | |
| US7782955B2 (en) | Transcoder and transcoding method operating in a transform domain for video coding schemes possessing different transform kernels | |
| CN101247525B (en) | Method for improving image intraframe coding velocity | |
| CN107396102B (en) | A kind of inter-frame mode fast selecting method and device based on Merge technological movement vector | |
| CN101022555B (en) | Interframe predictive coding mode quick selecting method | |
| HK1148888A1 (en) | Improved interpolation of video compression frames | |
| CN102932642A (en) | Interframe coding quick mode selection method | |
| CN106210721A (en) | A kind of HEVC quick code check code-transferring method | |
| CN102196272B (en) | P frame encoding method and device | |
| CN104519367A (en) | Video decoding processing apparatus and operating method thereof | |
| CN105611301B (en) | Distributed video decoding method based on wavelet field residual error | |
| CN104702959B (en) | A kind of intra-frame prediction method and system of Video coding | |
| JPH08265765A (en) | Image coding system and motion compensating device for use therein | |
| KR20080108538A (en) | Video coding | |
| CN101621689A (en) | MPEG-to-H.264/AVC video coding conversion system | |
| CN106131573B (en) | A kind of HEVC spatial resolutions code-transferring method | |
| CN100586186C (en) | Quick inter-frame forecast mode selection method | |
| CN101783956A (en) | Back-prediction forecast method based on spatio-temporal neighbor information | |
| CN100586185C (en) | Mode selection method for transcoding 264 video to reduce resolving capability | |
| KR100359819B1 (en) | An Efficient Edge Prediction Methods In Spatial Domain Of Video Coding | |
| CN101262607B (en) | Two-folded prediction video coding and decoding method and device | |
| KR101691380B1 (en) | Dct based subpixel accuracy motion estimation utilizing shifting matrix | |
| KR100899658B1 (en) | System and method for MPEG2/H.264 high speed digital video transcoding | |
| CN101977317A (en) | Intra-frame prediction 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090722 Termination date: 20130726 |