CN103096083B - A kind of moving image compensation method and device - Google Patents
A kind of moving image compensation method and device Download PDFInfo
- Publication number
- CN103096083B CN103096083B CN201310024458.6A CN201310024458A CN103096083B CN 103096083 B CN103096083 B CN 103096083B CN 201310024458 A CN201310024458 A CN 201310024458A CN 103096083 B CN103096083 B CN 103096083B
- Authority
- CN
- China
- Prior art keywords
- macro block
- frame
- image
- field
- compensation
- 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
Abstract
The invention discloses a kind of moving image compensation method and device, not fogging clear in compensating to solve moving image in prior art, the problem that picture quality is not high.In the present invention, the current frame image data of decoding moving picture, obtain motion compensation information, and the motion compensation information and previous frame image that obtain according to decoding, determine the reference frame of current frame image; Then corresponding to this reference frame view data is taken exercises compensation calculation, obtains image compensation data; The image compensation data that last basis obtains, carries out image compensation to current frame image, is compensated image, without the need to carrying out copying of data, can not image be overlapped, and the transitional function between adjacent two frame moving images can be played, make motion picture become more clear, improve picture quality.
Description
Technical field
The present invention relates to moving image compensation technique field, particularly relate to a kind of moving image compensation method and device.
Background technology
Growing along with liquid crystal technology, the appearance of the panel product of oversize and ultrahigh resolution, the requirement in picture quality is also more and more higher.
Generally, by the MEMC motion compensating module in drive circuit, process is compensated to moving image to the motion picture of TFT-LCD liquid crystal display in prior art.MEMC(MotionEstimateandMotionCompensation, Motion estimation and compensation) moving image compensates and adopts the method for frequency multiplication, the picture data of acquisition is copied portion, obtain frequency multiplication picture data, and utilize this frequency multiplication picture data to generate frame compensation data, then these frame compensation data are utilized to carry out image compensation, when namely adopting frequency doubling technology to carry out image compensation, picture data need be copied portion, then transmission process is carried out, therefore, when carrying out Data Stream Processing, within the same time, volume of transmitted data is exactly double, and adopt the mode of copy data, false image can be caused, None-identified True Data and the data after copying when carrying out image real time transfer, have superimposed image to occur.
Therefore, adopt the method that existing MEMC moving image compensating module processes image, the appearance due to superimposed image makes picture quality not high, and transmits data procedures complexity, and cause the drive circuit design of transmission data complicated, cost is higher.
Summary of the invention
The object of this invention is to provide a kind of moving image compensation method and device, not fogging clear in compensating to solve moving image in prior art, the problem that picture quality is not high.
The object of the invention is to be achieved through the following technical solutions:
One aspect of the present invention provides a kind of moving image compensation method, and the method comprises:
The current frame image data of decoding moving picture, obtain motion compensation information;
The described motion compensation information and previous frame image that obtain according to decoding, determine the reference frame of current frame image;
The view data corresponding to described reference frame carries out motion compensation calculations, obtains image compensation data;
According to described image compensation data, current frame image is compensated, be compensated image.
The present invention additionally provides a kind of moving image compensation arrangement on the other hand, and this device comprises:
Reference frame determination module, for the current frame image data of decoding moving picture, obtains motion compensation information, and the described motion compensation information and previous frame image that obtain according to decoding, determine the reference frame of current frame image;
Motion compensation calculations module, the view data that the reference frame for determining described reference frame determination module is corresponding carries out motion compensation calculations, obtains image compensation data;
Image compensation module, for the described image compensation data obtained according to motion compensation calculations module, carries out image compensation to current frame image, is compensated image.
Moving image compensation method provided by the invention and device, by calculating the reference frame of a present frame between consecutive frame image, and the view data corresponding to this reference frame does corresponding motion compensation calculations, obtain image compensation data, carry out image compensation, thus a transitional function is played between adjacent two frame motion pictures, and make motion picture become more clear, improve picture quality.
Accompanying drawing explanation
Fig. 1 is moving image compensation method realization flow figure provided by the invention;
The frame prediction reference macroblock positioning schematic diagram that Fig. 2 provides for the embodiment of the present invention;
The field prediction reference macroblock location schematic diagram that Fig. 3 provides for the embodiment of the present invention;
Fig. 4 predicts half precision interpolation schematic diagram for the frame that the embodiment of the present invention provides;
The field prediction half precision interpolation schematic diagram that Fig. 5 provides for the embodiment of the present invention;
The moving image compensation arrangement that Fig. 6 provides for the embodiment of the present invention forms schematic diagram.
Embodiment
Moving image compensation method provided by the invention and device, according to the motion compensation information obtained of decoding, determine the reference frame data of current frame image in previous frame image; Then corresponding moving image compensation calculation is done to this reference frame data, obtain image compensation data; The image compensation data reconstruct current frame image that last basis obtains, image compensation is carried out to current frame image, without the need to carrying out copying of data, the transmission of data volume can be reduced, and the reference frame that calculating one is virtual between consecutive frame carries out image compensation, the transitional function between adjacent two frame moving images can be played, and make motion picture become more clear, improve picture quality.
Below with reference to drawings and the specific embodiments, the present invention will be further described in detail, certainly do not regard it as and be limited.
As shown in Figure 1, be the realization flow figure of the moving image compensation that the embodiment of the present invention one provides, comprise:
Step S101: decoding current frame image, according to the motion compensation information obtained of decoding, determines the reference frame data of current frame image in previous frame image.
Concrete, image is carrying out in the process compressed, the data code flow relevant to image compensation is often adopted to carry out the expression of relevant information, after current frame image being decoded in the embodiment of the present invention, the motion compensation information relevant to image compensation can be obtained, the luminance color component form of such as image, the I of image P category-B type and field, top, the type etc. of field, the end and frame, the motion compensation information obtained according to these decodings and the view data of previous frame image, between the previous frame image that then can be adjacent at present frame, determine a reference frame relevant to current frame image.
Step S102: corresponding motion compensation calculations is done to reference frame data, obtains image compensation data.
Concrete, by after calculating the virtual reference frame image data between adjacent two frames in the embodiment of the present invention, corresponding motion compensation calculations can be done to this reference frame data again according to actual conditions, obtain final image compensation data.
Step S103: according to the image compensation data obtained, carries out image compensation to current frame image, is compensated image.
Concrete, when carrying out image compensation in the embodiment of the present invention, existing compensation method can be adopted, just the image compensation data of compensation image picture be to consecutive frame between reference frame data do the image compensation data that corresponding motion compensation calculations obtains.
The moving image compensation method that the embodiment of the present invention provides, by utilizing the reference frame calculated between consecutive frame, and the view data corresponding to this reference frame carries out motion compensation calculations, after obtaining image compensation data, moving image is compensated, and carry out Image Reconstruction, without the need to carrying out copying of data, also there will not be superimposed image, make motion picture become more clear, improve picture quality.
The embodiment of the present invention two, by conjunction with the determination of practical application to reference frame data in the moving image compensation method in embodiment one, is described in detail.
When determining the reference frame of current frame image in previous frame image in the embodiment of the present invention, can in units of macro block, current frame image is decoded, determine the initial reference macro block that each macro block is corresponding in former frame, according to macro block position and the macro block mode of each initial reference macro block determined, determine a final reference macro block corresponding with initial reference macro block respectively, be made up of the reference frame of present frame the final reference macro block determined.
In each two field picture, the form that several macroblock arrangement are in blocks, video coding algorithm is in units of macro block, macro block is encoded one by one, be organized into continuous print video code flow, therefore in the embodiment of the present invention in order to improve the accuracy of the image compensation data determined, the determination of image compensation reference frame data can be carried out in units of macro block.
Macro block is made up of a luminance pixel and two additional chroma pixel blocks, in general, luminance block is the block of pixels of 16x16 size, and the size of two colourity block of image pixels is determined according to the luminance color component form of its image, as for YUV420 sampled images, macro block is then the block of pixels of 8*8 size.Therefore, in order to the position of Obtaining Accurate macro block, the luminance color component form of image need be pre-determined.
After determining the position of current macro, in order to final reference macro block corresponding with this current macro in former frame accurately can be determined, also need the macro block mode determining initial reference macro block, a series of motion compensation information can be obtained, the value of such as motion compensation kind, macro block characteristics parameter and quantization parameter after macro block mode (Macroblock_modes) decoding.To decode the macro block characteristics parameter obtained according to macro block mode (Macroblock_modes), the motion vector that current macro is used and frame field prediction type can be determined.Relevant decoding length changeable code table corresponding when preferably can carry out Variable Length Code according to macro block data in the embodiment of the present invention, decode the motion vector used by current macro and frame field prediction type, and forward motion vector or backward motion vectors can be determined, frame prediction or field prediction, being field, top or field, the end, is first vector or second vector etc.
Certainly, after determining frame field prediction type, the motion vector that can obtain according to the frame field prediction type determined and decoding, determines required final reference macro block in former frame.
The embodiment of the present invention three is described in detail in conjunction with the determination of practical application to reference macroblock, does not certainly regard it as and is limited.
In the embodiment of the present invention, when frame field prediction type is frame prediction, final reference macro block is determined in former frame, specifically comprise: at the initial reference macro block position place of the corresponding current macro of previous frame image, a range of movement is determined according to motion vector, a macro block maximum with current macro similarity is determined in described range of movement, and using macro block maximum for this similarity as required final reference macro block.
The method that being illustrated in figure 2 the employing embodiment of the present invention provides carries out the schematic diagram that macro block is determined, the initial reference macro block that current macro A is corresponding in former frame is A1, if use A1 directly to determine the data value of A, error may be larger, because the A1 in previous frame image is for the A in current frame image, due to moving image especially High-motion picture or image associated scenario, the change of background, the similarity degree of image is not necessarily minimum, even may be very large, but, the time interval between two images is very short, the correlation of certain quantity in statistical significance is at least there is between A1 and A, so, following method is designed: directly do not predict A with A1 in the present invention, but according within the scope of certain around A1 of the direction of motion vector by all kinds of algorithm, find a final reference macro block B, make the view data similarity of B and A maximum, to be made up the deficiency of A1 by the B around A1, to reach the object of compensation, in Fig. 2, the current location of A is (X, Y), according to motion vector (a, b) direction, the A1 horizontal direction corresponding with A is moved a unit, vertical direction moves b unit, in the embodiment of the present invention, a and b is odd number, adopt the translation mode in Fig. 2, then can obtain final macro block B(X+a, Y+b).
When frame field prediction type is field prediction, in former frame, determines final reference macro block, specifically comprise:
At the initial reference macro block position place of the corresponding current macro of previous frame image, top field range of movement and field, end range of movement is determined respectively according to the motion vector of initial reference macro block, and a top field macro block maximum with current macro similarity is determined in the range of movement of field, top, determine in the range of movement of field, the end one with maximum field, the end macro block of current macro similarity, using top field macro block and field, end macro block as required final reference macro block.
The field prediction of frame data, field, top and field, the end is needed to predict respectively, as shown in Figure 3, A(X, Y) field, top needs with B(X+a, Y+b) the top field prediction in module, the field, the end of A needs with C(X+c, Y+d) field, the end is predicted, here the corresponding relation pushing up field and field, the end must meet, B and C is equally that the initial reference macro block A1 that A is corresponding gets according to image vector translation in former frame, in the embodiment of the present invention, image vector is (a, b) with (c, d), according to a, b, the parity of c and d and positive and negative translation is carried out to A1, in Fig. 3, a>0 is odd number, b<0 is odd number, c<0 is odd number, d>0 is even number.
Preferably, in the embodiment of the present invention when the view data corresponding to reference frame carries out motion compensation calculations, half precision pixels interpolation calculation can be carried out respectively to the view data of each reference macroblock determined, obtain the image compensation data of each corresponding macro block in present frame.Picture element interpolation, it is a kind of imaging mode that can increase valid pixel (actual pixels) when calculating pixel, namely half precision pixels interpolation refers to quantity and the density of pixel needed for reality, a kind of computational methods that simple average calculates required pixel value are carried out to original pixels, after obtaining required pixel value after half precision pixels interpolation calculation, according to the pixel of real image, and calculate according to certain compute mode, produce the pixel made new advances, and be inserted into the contiguous gap of original pixel, thus can realize increasing pixel total amount and the object increasing picture element density.
Predict the reference macroblock determined in the embodiment of the present invention with frame in Fig. 2 and Fig. 3, and the reference macroblock that field prediction is determined is illustrated respectively, does not certainly regard it as and be limited.
As shown in Figure 4, for frame in the embodiment of the present invention predicts half precision interpolation schematic diagram, in Fig. 2, image vector a and b is odd number, namely when carrying out half precision interpolation to reference macroblock B according to parity, need to have more a line on the horizontal scale, if macro block A is the macro block of 16*16 size, then reference macroblock B is the macro block of 17*17 size, when carrying out parity half precision interpolation, the mean value that simple arithmetic average can be utilized to obtain is as the image compensation data of macro block A, macro block A after being compensated, as shown in Figure 5, a line had more in Fig. 5 and one is listed in averaging process and is cancelled, so the shape of image can not be changed.
As shown in Figure 5, for field prediction in the embodiment of the present invention half precision interpolation schematic diagram, in Fig. 3, the field, top of A needs to determine with the field, top of reference macroblock B, field, the end needs to determine with the field, the end of reference macroblock C, in Fig. 3, a>0 is odd number, b<0 is odd number, c<0 is odd number, d>0 is even number, according to image vector (a, b) the reference macroblock B found, because a and b is odd number, so B will do half precision interpolation in level and vertical both direction, and utilize image vector and (c, d) the reference macroblock C found, c is only had to be odd number, so only half precision interpolation need be done in the horizontal direction when C macro block does half precision interpolation.
Be still that the macro block of 16*16 size is described for example with A in the embodiment of the present invention, in Fig. 5, why reference macroblock B is 18*17, is that horizontal direction is identical with frame because for field, if will do half precision interpolation, must calculate one, i.e. 17 row more; And in the vertical direction, because adjacent two interlines in field are every a line, so must add two row by the vertical direction, i.e. 18 row altogether, when doing interpolation, the position that what a0, a1, a2 and a3 used is all in field, i.e. necessary interlacing.For C macro block, do interpolation because only have in horizontal direction, so be laterally 17 row, and longitudinally constant, so longitudinally remain 16 row.
By carrying out half precision interpolation respectively to the reference macroblock data in frame prediction and field prediction in the embodiment of the present invention, corresponding offset data can be obtained respectively for each macro block, make follow-up when carrying out image compensation, accuracy is higher, makes picture display effect better.
The embodiment of the present invention four additionally provides a kind of moving image compensation arrangement, and be illustrated in figure 6 the moving image compensation arrangement schematic diagram that the embodiment of the present invention provides, this device comprises:
Reference frame determination module 61, for the current frame image data of decoding moving picture, obtains motion compensation information, and the motion compensation information and previous frame image that obtain according to decoding, determine the reference frame of current frame image;
Motion compensation calculations module 62, the view data that the reference frame for determining reference frame determination module 61 is corresponding is taken exercises compensation calculation, obtains image compensation data;
Image compensation module 63, for the image compensation data obtained according to motion compensation calculations module 62, carries out image compensation to current frame image, is compensated image.
Preferably, reference frame determination module 61, specifically for:
In units of macro block, current frame image is decoded, obtain the motion compensation information of each macro block in present frame; According to the motion compensation information of each macro block obtained of decoding, determine the initial reference macro block that in present frame, each macro block is corresponding in previous frame image; According to macro block position and the macro block mode of each initial reference macro block, determine the final reference macro block corresponding with initial reference macro block respectively, be made up of the reference frame of present frame some final reference macro blocks determined.
Wherein, reference frame determination module 61, also for:
According to macro block position and the macro block mode of initial reference macro block, determine the motion vector that initial reference macro block is used and frame field prediction type; Wherein, motion vector comprises forward motion vector and backward motion vectors, and frame field prediction type comprises frame prediction and field prediction;
According to the motion vector that the frame field prediction type determined and decoding obtain, in former frame, determine final reference macro block.
Further preferred, reference frame determination module 61, also for:
When frame field prediction type is frame prediction, at the initial reference macro block position place of the corresponding current macro of previous frame image, a range of movement is determined according to motion vector, a macro block maximum with current macro similarity is determined in the range of movement determined, and using macro block maximum for this similarity as required final reference macro block;
When described frame field prediction type is field prediction, at the initial reference macro block position place of the corresponding current macro of previous frame image, top field range of movement and field, end range of movement is determined respectively according to motion vector, and a top field macro block maximum with current macro similarity is determined in the range of movement of field, top, determine in the range of movement of field, the end one with maximum field, the end macro block of current macro similarity, using top field macro block and field, end macro block as required final reference macro block.
Preferably, motion compensation calculations module 62, also for:
According to the parity of motion vector, the view data corresponding to each final reference macro block does half precision pixels interpolation calculation respectively, obtains the image compensation data with the corresponding macro block of each final reference macro block in present frame.
It should be noted that, the moving image compensation method involved by embodiment one to embodiment three, all can the moving image compensation arrangement that provides in the embodiment of the present invention of connected applications, do not repeat them here.
The moving image compensation arrangement that the embodiment of the present invention provides, by calculating a reference frame data between consecutive frame image, and half corresponding precision interpolation calculation is done to this reference frame data, obtain image compensation data, carry out image compensation, thus a transitional function is played between adjacent two frame motion pictures, and make motion picture become more clear, improve picture quality.
Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (6)
1. a moving image compensation method, is characterized in that, the method comprises:
In units of macro block, current frame image is decoded, obtain the motion compensation information of each macro block in present frame;
According to the motion compensation information of each macro block obtained of decoding, determine the initial reference macro block that in present frame, each macro block is corresponding in previous frame image;
According to macro block position and the macro block mode of described initial reference macro block, determine the motion vector that described initial reference macro block is used and frame field prediction type; Described motion vector comprises forward motion vector and backward motion vectors, and described frame field prediction type comprises frame prediction and field prediction;
According to the motion vector that the frame field prediction type determined and decoding obtain, determine final reference macro block, be made up of the reference frame of present frame some described final reference macro blocks;
The described motion compensation information and previous frame image that obtain according to decoding, determine the reference frame of current frame image;
The view data corresponding to described reference frame carries out motion compensation calculations, obtains image compensation data;
According to described image compensation data, current frame image is compensated, be compensated image.
2. the method for claim 1, is characterized in that, the motion vector that described frame field prediction type according to determining and decoding obtain, and determines described final reference macro block, specifically comprises:
When described frame field prediction type is frame prediction, at the initial reference macro block position place of the corresponding current macro of previous frame image, a range of movement is determined according to described motion vector, a macro block maximum with current macro similarity is determined in described range of movement, and using macro block maximum for this similarity as described final reference macro block;
When described frame field prediction type is field prediction, at the initial reference macro block position place of the corresponding current macro of previous frame image, top field range of movement and field, end range of movement is determined respectively according to described motion vector, and a top field macro block maximum with current macro similarity is determined in the range of movement of field, top, determine in the range of movement of field, the end one with maximum field, the end macro block of current macro similarity, using described top field macro block and field macro block of the described end as described final reference macro block.
3. the method as described in any one of claim 1-2, is characterized in that, the view data corresponding to described reference frame carries out motion compensation calculations, obtains image compensation data process, specifically comprises:
According to the parity of motion vector, the view data corresponding to each final reference macro block does half precision pixels interpolation calculation respectively, obtains the image compensation data with the corresponding macro block of each final reference macro block in present frame.
4. a moving image compensation arrangement, is characterized in that, this device comprises:
Reference frame determination module, in units of macro block, decodes to current frame image, obtains the motion compensation information of each macro block in present frame; According to the motion compensation information of each macro block obtained of decoding, determine the initial reference macro block that in present frame, each macro block is corresponding in previous frame image; According to macro block position and the macro block mode of described initial reference macro block, determine the motion vector that described initial reference macro block is used and frame field prediction type; Described motion vector comprises forward motion vector and backward motion vectors, and described frame field prediction type comprises frame prediction and field prediction; According to the motion vector that the frame field prediction type determined and decoding obtain, determine final reference macro block, be made up of the reference frame of present frame some described final reference macro blocks;
Motion compensation calculations module, the view data that the reference frame for determining described reference frame determination module is corresponding carries out motion compensation calculations, obtains image compensation data;
Image compensation module, for the described image compensation data obtained according to motion compensation calculations module, carries out image compensation to current frame image, is compensated image.
5. moving image compensation arrangement as claimed in claim 4, is characterized in that, described reference frame determination module, also for:
When described frame field prediction type is frame prediction, at the initial reference macro block position place of the corresponding current macro of previous frame image, a range of movement is determined according to described motion vector, a macro block maximum with current macro similarity is determined in described range of movement, and using macro block maximum for this similarity as described final reference macro block;
When described frame field prediction type is field prediction, at the initial reference macro block position place of the corresponding current macro of previous frame image, top field range of movement and field, end range of movement is determined respectively according to described motion vector, and a top field macro block maximum with current macro similarity is determined in the range of movement of field, top, determine in the range of movement of field, the end one with maximum field, the end macro block of current macro similarity, using described top field macro block and field macro block of the described end as described final reference macro block.
6. the moving image compensation arrangement as described in any one of claim 4-5, is characterized in that, described motion compensation calculations module, specifically for:
According to the parity of motion vector, the view data corresponding to each final reference macro block does half precision pixels interpolation calculation respectively, obtains the image compensation data with the corresponding macro block of each final reference macro block in present frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310024458.6A CN103096083B (en) | 2013-01-23 | 2013-01-23 | A kind of moving image compensation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310024458.6A CN103096083B (en) | 2013-01-23 | 2013-01-23 | A kind of moving image compensation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103096083A CN103096083A (en) | 2013-05-08 |
| CN103096083B true CN103096083B (en) | 2015-12-09 |
Family
ID=48208146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310024458.6A Expired - Fee Related CN103096083B (en) | 2013-01-23 | 2013-01-23 | A kind of moving image compensation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103096083B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10725287B2 (en) * | 2013-06-11 | 2020-07-28 | Nlight, Inc. | Image rotation compensation for multiple beam material processing |
| CN106652002A (en) * | 2016-09-20 | 2017-05-10 | 努比亚技术有限公司 | Method and system for generating trajectory animation image |
| CN112307968B (en) * | 2020-10-30 | 2022-11-08 | 天地伟业技术有限公司 | Face recognition feature compression method |
| CN112906475B (en) * | 2021-01-19 | 2022-08-02 | 郑州凯闻电子科技有限公司 | Artificial intelligence-based rolling shutter imaging method and system for urban surveying and mapping unmanned aerial vehicle |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101163244A (en) * | 2007-11-22 | 2008-04-16 | 上海交通大学 | Picture element caching method and system in movement compensation process of video decoder |
| CN102223525A (en) * | 2010-04-13 | 2011-10-19 | 富士通株式会社 | Video decoding method and system |
| US20110255597A1 (en) * | 2010-04-18 | 2011-10-20 | Tomonobu Mihara | Method and System for Reducing Flicker Artifacts |
| CN102340662A (en) * | 2010-07-22 | 2012-02-01 | 炬才微电子（深圳）有限公司 | Video processing device and method |
-
2013
- 2013-01-23 CN CN201310024458.6A patent/CN103096083B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101163244A (en) * | 2007-11-22 | 2008-04-16 | 上海交通大学 | Picture element caching method and system in movement compensation process of video decoder |
| CN102223525A (en) * | 2010-04-13 | 2011-10-19 | 富士通株式会社 | Video decoding method and system |
| US20110255597A1 (en) * | 2010-04-18 | 2011-10-20 | Tomonobu Mihara | Method and System for Reducing Flicker Artifacts |
| CN102340662A (en) * | 2010-07-22 | 2012-02-01 | 炬才微电子（深圳）有限公司 | Video processing device and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103096083A (en) | 2013-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7313816B2 (en) | Image prediction method and related device | |
| US8385628B2 (en) | Image encoding and decoding method, apparatuses therefor, programs therefor, and storage media for storing the programs | |
| CN110741640B (en) | Optical flow estimation for motion compensated prediction in video coding | |
| JP2021528011A (en) | Subblock motion vector inheritance between color components | |
| CN108141606B (en) | Method and system for global motion estimation and compensation | |
| US20120314772A1 (en) | Optical flow based motion vector estimation systems and methods | |
| EP1406449A2 (en) | Interpolation apparatus and method for motion compensation | |
| KR20150018576A (en) | Motion compensation and motion estimation leveraging a continuous coordinate system | |
| CN103096083B (en) | A kind of moving image compensation method and device | |
| US8068542B2 (en) | Motion picture encoding device and motion picture decoding device | |
| US20210211679A1 (en) | Non-affine blocks predicted from affine motion | |
| US20150110190A1 (en) | Method and apparatus for motion estimation | |
| US10284810B1 (en) | Using low-resolution frames to increase frame rate of high-resolution frames | |
| CN1965583A (en) | Method of searching for a global motion vector | |
| CN103929648A (en) | Motion estimation method and device in frame rate up conversion | |
| AU2021298606B2 (en) | Encoding and decoding method and apparatus, and device therefor | |
| CN110692241A (en) | Diversified motion using multiple global motion models | |
| CN111010571A (en) | Generation and use of combined affine Merge candidates | |
| CN101242530B (en) | Motion estimation method and multi-video coding and decoding method and device based on motion estimation | |
| US20190141332A1 (en) | Use of synthetic frames in video coding | |
| KR100254667B1 (en) | Method and apparatus for detecting occulusion | |
| CN103313059B (en) | The decision method of occlusion areas during a kind of frame per second promotes | |
| JP2009164865A (en) | Video coding method, video decoding method, video coding apparatus, video decoding apparatus, programs therefor and computer-readable recording medium | |
| AU2021286043B2 (en) | Encoding and decoding method and apparatus, and device therefor | |
| Joshi et al. | VLSI architecture of high speed SAD for high efficiency video coding (HEVC) encoder |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151209 Termination date: 20210123 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |