CN1756350A - Motion image coding method - Google Patents
Motion image coding method Download PDFInfo
- Publication number
- CN1756350A CN1756350A CN 200410051733 CN200410051733A CN1756350A CN 1756350 A CN1756350 A CN 1756350A CN 200410051733 CN200410051733 CN 200410051733 CN 200410051733 A CN200410051733 A CN 200410051733A CN 1756350 A CN1756350 A CN 1756350A
- Authority
- CN
- China
- Prior art keywords
- macro block
- synchronizing information
- piece group
- data
- hidden
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention relates to dynamic image coding method, which comprises: to input image, extracting synchronous information of some macroblock formed block group unit; to the extracted information, transmitting with data hiding when coding; extracting said information with data hiding from decoder to form macroblock synchronous information and execute error hiding. Wherein, the said synchronous information relates to the initial position of sixth macroblock on bit flow. Besides, from this invention, using grade value by running quantization parameter and /or DCT to hide data.
Description
[technical field]
The present invention is the invention of relevant moving image coding, when particularly relevant multi-media image passes through Network Transmission, by the data hidden of cataloged procedure, to the mistake that decode procedure takes place, can effectively handle the moving video encoding method of hiding error.
[background technology]
Recently,, be not only audio frequency and still image, and the multimedia service of moving image and so on enlarges also along with international mobile communication 2000 ultrahigh speed mobile communication such as (IMT2000) gets a day practical application.As the example of moving image multimedia service, the bi-directional service of the one directional service of video request program (VOD) and so on and video call and so on is arranged.Obviously more because of video communication than the voice-frequency telephony data volume, move video compression basically.
It is such that general moving image coding process and accompanying drawing 1 are represented, to input picture, extract motion vector by dynamic compensation (Motion compensation) and dynamic estimation (Motion estimation), and to its error image operation discrete cosine transform and quantization process.For to moving quantized data, processing data compression more effectively, operation variable length code (VLC Encoding).
In addition, operation anti-quantization, inverse discrete cosine transformation and before this image lump together, be reduced to image before this, and obtain the error image of next image with this image.Finish video data compression by this way.
Through after such video compression, converse by network and the other side.At this moment, the network generation noise that moves of data.That is, the data of communication are not to be transferred to receiving terminal faithfully, but corrupted data or disappearance take place.The impaired of such transmission data or disappearance are called mistake.
Such mistake not only exerts an influence to the frame that makes a mistake, and also exerts an influence to continuity afterwards.Its reason is that because of video is moved compression, the dynamic estimation of space-time makes the once wrong data of generation exert an influence to successional frame.In addition, when error in data took place, the variable length code mode lost (Sync) synchronously, and it is exerted an influence till finding the next one synchronous at that frame.That is,,, and make its continuous appearance to low the change greatly of image quality of data loss because of video compression.
Particularly, when data compression made a mistake, till finding next synchronizing information, all incorrect because of all information, if the frequent more appearance of synchronizing information, error propagation is just possible more significantly to be reduced.But because of synchronizing information is different with the bit stream of other coding, message length is longer, can not frequently enter video compression.Such as, H.263, have only picture top (Picture header) and front, piece group top to comprise synchronizing information.Because of compressive flow is not often to comprise synchronizing information, if make a mistake, till next synchronizing information occurring, all corrupted data make the low change of image quality big.
The above low situation of image quality is the big obstacle of image service.At video compression standard, also have the reduction of such mistake and hidden method, but certain limit is all arranged.
Method of reducing to mistake mainly is a method of inserting supplemental data with encoder, when making a mistake, is intended to the method that prevents that decoder error from propagating.This is that a kind of appending on the stream of coding inserted sync bit or the stream of coding is divided into equal length, prevents the method for error propagation.There is such weakness in wrong method of reducing like this,, if decoder is not known the wrong method of reducing of encoder, just can not move decoding that is.
The method that another one method in the mistake method of reducing has encoder to move simultaneously.This method is to utilize the method for backward channel (back channel), if decoder shows the position of making a mistake to encoder, encoder just utilizes this information, encodes with the method that can carry out wrong reduction.At this moment, have such weakness, that is, the support backward channel should append in system.
Error hidding method is the method for carrying out at decoder, and it is the image that utilizes in the decoder, and the part that makes a mistake is corrected as method with original image similarity to greatest extent.At this moment, because of utilize the hidden mistake of finite information at decoder, there is certain limit in the correctness of hiding error.
Be all to remove the mistake of generation only with the wrong method of reducing of above explanation and the mode of error hidding method.
[summary of the invention]
The objective of the invention is to, when multi-media image passes through Network Transmission, provide data hidden, form certain macro block synchronizing information of piece group unit by cataloged procedure, to the mistake that cataloged procedure takes place, provide the moving video encoding method that effectively to handle hiding error with this.
To achieve these goals, moving video encoding method of the present invention is relevant to the moving hidden moving video encoding method of image transmission system run-time error, and the method comprises the steps to be characteristics:
To the image of encoder input, extract the step of the certain macro block synchronizing information that forms piece group unit;
To certain macro block synchronizing information of the formation piece group unit of said extracted, the data hidden during by coding, the step of transmitting;
Extract according to above-mentioned data hidden transmission from decoder, and form certain macro block synchronizing information of piece group unit, also utilize the synchronizing information of extracting, the step that run-time error is hidden.
, according to explanation of the present invention, have following characteristics here: the certain macro block synchronizing information that forms above-mentioned group unit is meant that with the piece group be unit, the 6th macro block starting position on bit stream;
To certain macro block synchronizing information of the piece group unit that forms said extracted, when service data is hidden, hide certain macro block synchronizing information of the piece group unit of formation said extracted repeatedly;
When moving above-mentioned data hidden, utilize the grade point (removing the value of discrete cosine transform coefficient with the quantization parameter) in the piece of input picture operation quantization parameter and/or discrete cosine transform, service data is hidden;
Certain macro block synchronizing information to the piece group unit that forms said extracted when service data is hidden, is hidden in certain macro block synchronizing information of said extracted a piece group of the piece group back of the certain macro block synchronizing information of said extracted;
Certain macro block synchronizing information that utilization is extracted from above-mentioned decoder is when run-time error is hidden, if the piece group macro block of input makes a mistake, just with reference to extracting from hiding data, and the macro block synchronizing information of formation piece group, the macro block to the above-mentioned piece group that makes a mistake reruns decoding;
When certain macro block synchronizing information of extracting from above-mentioned decoder belongs to the synchronizing information of the 6th macro block that forms a piece group, if the above-mentioned macro block that makes a mistake is the macro block before the 6th macro block, just, begin to rerun decoding from the 6th macro block of the above-mentioned piece group that makes a mistake with reference to the synchronizing information of said extracted.
As described above, according to moving video encoding method of the present invention, there is such strong point: when multi-media image passes through Network Transmission, data hidden by cataloged procedure, the certain macro block synchronizing information that forms piece group unit is provided, to the mistake that decode procedure takes place, can effectively handle hiding error with this.
[description of drawings]
Accompanying drawing 1 is roughly represented the process according to general moving video encoding method operation coding.
Accompanying drawing 2 is the drawings for the image stratum that general moving video encoding method is described.
Accompanying drawing 3 is at general moving video encoding method, the drawing of influence when illustrating that each image stratum makes a mistake.
Accompanying drawing 4 is for illustrating that moving video encoding method of the present invention provides the drawing of the synchronizing information effect of each piece group.
Accompanying drawing 5 is for illustrating that moving video encoding method of the present invention utilizes the quantization parameter, the drawing of the hiding process of service data.
Accompanying drawing 6 is for illustrating that moving video encoding method of the present invention utilizes grade point, the drawing of the hiding process of service data.
Accompanying drawing 7 is expression moving video encoding method of the present invention precedence diagrams at encoder operation cataloged procedure.
Accompanying drawing 8 is expression moving video encoding method of the present invention precedence diagrams at decoder operation cataloged procedure
[embodiment]
Below, with reference to additional drawing, describe embodiments of the invention in detail.
The present invention, hides the certain macro block synchronizing information that forms piece group unit, and transmits by data hidden at cataloged procedure, when providing decoder to make a mistake with this, more effectively handles the moving video encoding method of hiding error.
Known from institute, the video information of video communication is more because of data volume, transmits after the operation compression.If whole compressed video datas, the data volume of compression also increases, and therefore, finds out and the part of image similarity before this, obtain motion vector, and behind this a part of error image of encoding, the operation variable length code is transmitted.
If with such method compressed video data, just can carry out more data compression, but when making a mistake, propagation makes a mistake.That is, because of using the variable length code compression, even make a mistake in a position, its mistake continues till the next synchronizing symbol up to occurring, and because of with reference to image encoding before this, if image makes a mistake before this, its error propagation is to next image.
In addition, the general video-frequency compression method that compresses by dynamic estimation and dynamic compensation is image segmentation a stratum, and utilizes stratum's operation compression.Utilize stratum to be because can be included in top (header) to the common information of next stratum.The stratum of common dynamic estimation and dynamic compensation video-frequency compression method-H.263 is illustrated in accompanying drawing 2, and accompanying drawing 2 is the drawings that generally move the image stratum of video encoding method for illustrating.
As shown in Figure 2, the stratum of image can be divided into picture layer, gob layer, macroblock layer, four stratum of piece layer.Here, picture is represented a frame, and comprises 9 piece groups.The piece group represents with the macro block to be row of unit, comprises 11 macro blocks.Macro block comprises four brightness (Luminance) pieces and two chrominance block as the unit that finds out motion vector.Piece comprises the 8*8 pixel as the base unit of data compression.
More detailed sight is wiped just can learn that each stratum all has the top, represents the structural information of stratum below it.The picture top comprise picture type of coding, append pattern use whether, the whole frame information of frame such as reference frame number, picture specification.Piece group top comprises that the piece group begins code and information such as piece sets of numbers, frame ID.That is, comprise information to the macro block being row of unit.The macro block top is a unit with the pixel of 16*16, comprises information such as motion vector, macroblock encoding type.
As mentioned above, each stratum has the top that comprises following hierarchical structure information.Under the video communication environment, when make a mistake in the top of each stratum, the variation when image quality lowly changes obviously greater than the generation error in data.Accompanying drawing 3 expressions when make a mistake in each stratum top, produce the low phenomenon of which kind of image quality, and its influence reach any degree.
As shown in Figure 3, topmost stratum-when make a mistake in the picture top, because of the entire frame of can not decoding, the image quality bigger with reference to the next frame generation of this frame is low.Particularly, when moving greatly, if fail reference image before this, image quality is low just worse.
In addition, when make a mistake in piece group top, row that can not proper solution code block group; When make a mistake in the macro block top, to a last macro block, all can not move normal decoder from the macro block that makes a mistake.
In addition, at decode procedure, picture top and piece group top are because of comprising synchronizing symbol, and coded data makes a mistake, and lose synchronously, in the time of can not normal decoder, if run into piece group top or picture top, just find out synchronously, and can carry out normal decoder again.But because of the macro block top does not comprise synchronizing symbol, the top or the data of the following unit of macro block make a mistake, because variable length code, and content that can not the normal decoder data.
So, till running into the picture top or piece group top that next synchronizing symbol occurs, lose all data.That is, such as shown in Figure 4 even the mistake that takes place is a simple bit-errors, the unlikely reduction of all signals till next synchronizing symbol occurring.
The present invention is low in order to prevent such image quality, utilizes the data hidden method, the certain macro block synchronizing information that forms piece group unit is hidden in the coded data transmits, with this error running image restoring that takes place will be to decoding the time.
Below, with the data hidden of explanation the present invention operation.
Data hidden be Information hiding at digital multimedia, and where necessary, extract the technology that hides Info.The technology of hiding data is broadly divided into, and for extraction hides Info, needs the technology of original paper and does not need the technology of original paper.The data-hiding method that the present invention uses is the method that does not need former event data.
Data hidden of the present invention is to move when video compression.Data hidden is when the compression input picture, and the parameter of utilizing during with the change data compression or the dependence value of input picture are moved.Want hiding data during the operation video compression, this some should be such part, that is, even carry out data hidden, original parameter or former event data change, and produces considerable influence also for the data volume of image quality or compression.Such value has, and the quantization parameter, removes the grade point of discrete cosine transform coefficient with the quantization parameter.
Utilizing the quantization parameter, the procedural representation that service data is hidden is at accompanying drawing 5.Accompanying drawing 5 is moving video encoding methods according to the present invention, and for explanation utilizes the quantization parameter, service data is hidden the drawing of process.
As everyone knows, adjusting coded data amount is the quantization parameter.The quantization parameter is to remove the parameter of the coefficient value of the difference part of input picture or input picture, discrete cosine transform (DCT).If it is big that the quantization parameter becomes, the value of removing the DCT coefficient will become greatly, therefore, needs the coded data amount to reduce; If the quantization parameter diminishes, the value of removing the DCT coefficient will diminish, and therefore, needs the coded data amount to increase.
In addition, when moving image, to be fit to the frequency bandwidth packed data of network by Network Transmission and reception.If the frequency bandwidth of network is wide, data volume that can communication just increases, and therefore, needs the coded data amount to increase, and makes image quality higher.On the contrary, if the frequency bandwidth of network is narrow, data volume that can communication just tails off, and therefore, needs the coded data quantitative change few, makes image quality low.Like this, according to network condition, utilize quantization parameter regulation coded data amount, at this moment service data is hidden.
As shown in Figure 5, behind the error image operation DCT to input picture or input picture, consider the frequency bandwidth of network, set suitable quantization parameter.The quantization parameter enters the top of macro block, is used for decoding compressed image.At this moment, utilize actual quantization parameter, before the operation quantization, service data is hidden.Here, data hidden can be according to following formula operation.
QP_new％2＝Hide?bit[k]
QP_new does not change
QP_new％2！＝Hide?bit[k]
QP_new＝QP_new+1；
Hide bit[k]: bit stream that will hiding data
As mentioned above, if service data is hidden the data of the decoder of just decoding compression, and according to the value of quantization parameter, can make it extract the data of hiding.If the quantization parameter of decoder belongs to even number, hiding data just shows with ' 0 '; If the quantization parameter belongs to odd number, hiding data just shows with ' 1 '.
At this moment, for data hidden, encoder changes the quantization parameter slightly.That is, when removing the data that to hide and quantization parameter,, make with 2 two remainders removing identical if its remainder difference just increases by 1 to the quantization parameter value with 2.Like this, because of the value of removing the DCT coefficient becomes big, need the coded data amount to reduce.But the image quality that can with the naked eye feel lowly can take place hardly.
In addition, to the situation that the value-grade point service data of the DCT coefficient value that moves quantized input picture or error image is hidden, represent with accompanying drawing 6.Accompanying drawing 6 is according to moving video encoding method of the present invention, and for explanation utilizes grade point, service data is hidden the drawing of process.
According to moving video encoding method of the present invention, in order to compress, to input picture or error image operation DCT.Then, consider the frequency bandwidth that will transmit, specify the quantization parameter, the position amount that adjusting will be encoded.The value that the quantization parameter is removed is the DCT coefficient.
At accompanying drawing 6, DCT is the unit operation with the 8*8 piece.The quantization parameter is suitable for the piece of operation DCT, and removes coefficient value, and at this moment, the share of removing with the quantization parameter is called grade.Each piece utilizes share-grade, and service data is hidden.Here, data hidden can be according to following formula operation.
Level does not change
In Level, the Level value minimum importance degree reduces 1
LevelSum: the value that all adds the Level of piece
Not to hide, but be unit, adding that the reason that service data is hidden on the value of grade is, when each grade service data is hidden, can exert an influence to image quality with the piece in separately Level (grade) service data.Because of grade is to remove the value of DCT coefficient with the quantization parameter, the quantization parameter is big more, and the scope of the DCT coefficient that 1 grade is represented is big more.
Such as, even grade point all equals 1, if the quantization parameter is 10, possible DCT scope is from 10 to 19; If the quantization parameter is 30, possible DCT scope is from 30 to 59.Therefore, be unit with the 8*8 piece, obtain the total (Sum) of grade after, if with 2 remove remainder that grade adds up to and will hiding data identical, just represent nonrated variation; If it is worth different, just the minimum grade point of importance degree is reduced 1.
Here, the grade that importance degree is minimum is meant the minimum part of the susceptibility of naked eyes.Because of people's eyes are lower to the susceptibility in high-frequency field, if with 2 remove will hiding data the position different with the remainder that adds up to of grade, just to having the grade point minimizing 1 of highest frequency.
That is, if grade adds up to 16, and the position that will insert is 1, just the grade point in the highest frequency field of its piece reduced 1, the grade of piece added up to become 15, makes with this that to remove the remainder of the position that will hide and grade total with 2 identical.
If but grade adds up to 1, just go wrong.Be 1 if grade adds up to the position that will hide, just the minimum grade of susceptibility reduced 1, make whole total become 0.But the total of original piece equals 0 piece because of not reducing grade, not hiding data again.Like this, just can not distinguish original grade and add up to 1, but behind the hiding data, grade adds up to and becomes 0 piece and original grade adds up to 0, and does not have the piece of data hidden.
The present invention has used following method in order to prevent such error.Grade adds up at 1 o'clock, if the data that will hide are 1, just inserts data at that piece.Even like this, the grade total does not change.If grade adds up to 1, and the data that will hide are 0, just the grade of its piece are added up to change to 0, do not carry out data hidden.Grade adds up at 0 o'clock, and decoder thinks that there is not data hidden in its piece.
With such method, except grade added up to 1 piece, if it is identical to remove the remainder that will hiding data adds up to self piece with 2, the grade of all pieces did not just change; Otherwise, just the minimum grade of susceptibility is reduced 1, can hiding data with this.
If the data that will hide are 1, grade adds up to 1 piece with regard to hiding data; If the data of hiding are 0, just hiding data not only changes to 0 to the grade point of piece.Grade adds up at 0 o'clock, and also service data is not hidden.
When the present invention utilizes quantization parameter or grade point service data to hide, the certain macro block synchronizing information service data that forms piece group unit is hidden.
Here, the certain macro block synchronizing information that forms piece group unit is meant the starting position on the bit stream of the 6th macro block in centre in 11 macro blocks that a piece group has.Such piece group intermediate position information is to be inserted in a piece group back transmission.
In more detail, on bit stream, the starting position transmission of hiding the 6th macro block is the position that is illustrated in each the 6th macro block on the bit stream, inserts synchronizing information.
As mentioned above, when bit stream makes a mistake, till running into next synchronizing symbol, can not all data of normal decoder.But according to the present invention, in the starting position that bit stream is uploaded defeated the 6th macro block, with this when the decoder decode process makes a mistake, can be aiming at the 6th macro block that is not piece group top synchronously, therefore, can obtain better image quality, and also can reduce wrong propagation.
Such as, as shown in Figure 4, if the 3rd macro block makes a mistake, method is till next piece group occurs before this, can not all data of normal decoder.But as of the present invention,, know the starting position of the 6th macro block if utilize hiding of synchronizing information, though can not normal decoder from the 3rd to the 5th macro block data, since the 6th macro block, can normal decoder.
In addition, considering from the encoder side that the synchronizing information service data is hidden, if see the data of wiping with the compression of piece group unit, because of operation variable length code compression, is unit with the piece group, and the position of the 6th piece group is different separately.So from being that the 6th piece group position extracted in the position of unit encoding with the piece group, the positional information that makes its extraction is inserted when encoding next piece group.At this moment, make the starting position of the 6th macro block be hidden in quantization parameter and discrete cosine transform grade point.This procedural representation at accompanying drawing 7.Accompanying drawing 7 is according to moving video encoding method of the present invention, with the precedence diagram of representation of concept encoder operation cataloged procedure.
In addition, the decoder side process of extracting synchronizing information is described, during the operation decoding, extracting with the piece group is the information that unit hides, and learns the starting position of the 6th macro block of piece group before this.In piece group before this, when making a mistake before the 6th the piece group, utilize the information of extracting, from the 6th macro block of piece group before this, normal again operation decoding.This process is represented with accompanying drawing 8.Accompanying drawing 8 is according to moving video encoding method of the present invention, the precedence diagram of expression decoder operation cataloged procedure.
In addition, in moving image communication, error rate changes according to environment.If serious, just reach 10E-3, existing does not almost have wrong environment yet.At this moment, at the higher environment of wrong incidence, the data that will hide are also destroyed because of mistake, therefore, answer the repeated multiple times transmission.When the data of reduction make a mistake, extract hiding data, hidden mistake.Its reason is, factor is according to hiding the bit stream that itself is hidden in coding, if coded data is destroyed by mistake, the data of hiding are also destroyed.
So if utilize destruction information, run-time error is hidden, image quality is lowly worse, therefore, repeatedly transmits the data that will hide repeatedly, when the hiding data of decoder destroys because of mistake, also makes its utilization other data repeatedly.
In addition, now the error hidding method of operation only utilizes the image that decoder receives, and makes it be reduced to image with original image similarity to greatest extent.That is, limited because of utilizable information, make and go back original image and have gap with original image.The error image of image compresses general video compression mode because of utilizing before this, when certain makes a mistake on a bit, if not exclusively remove that mistake partly, just might give next image propagates mistake.That is, need to remove to greatest extent the mistake that makes a mistake a little.
Utilize moving video encoding method of the present invention,, and receive also in transmission ends that to append the information run-time error hidden, can remove mistake to greatest extent because of not only hidden at the receiving terminal run-time error.If the algorithm of using the present invention to propose when making a mistake, not only can find synchronizing information at picture top and piece group top, and in the middle of the piece group, also synchronizing information can be found, therefore, error propagation can be significantly prevented.
If utilize moving video encoding method of the present invention, when video communication services such as IMT-2000 make a mistake, make hiding error have more validity.Like this, can be effective to use the moving image communication service of general video multimedia digital signal codec (CODEC) standard (H.26X, MPEG1.2.4).
Claims (7)
1, moving video encoding method, described method is at moving image transmission system, run-time error hidden moving video encoding method, comprise step:
To image, extract the certain macro block synchronizing information that forms piece group unit in the encoder input;
To certain macro block synchronizing information of the formation piece group unit of said extracted, the data hidden during by coding transmits;
Be extracted in decoder and transmit because of above-mentioned data hidden, and form certain macro block synchronizing information of piece group unit, also utilize the synchronizing information of extracting, run-time error is hidden.
2, moving video encoding method as claimed in claim 1 is characterized in that,
The certain macro block synchronizing information that forms above-mentioned group unit is meant that with the piece group be unit, the 6th macro block starting position on bit stream.
3, moving video encoding method as claimed in claim 1 is characterized in that,
To certain macro block synchronizing information of the piece group unit that forms said extracted, when service data is hidden, hide certain macro block synchronizing information of the piece group unit of formation said extracted with repeated multiple times.
4, moving video encoding method as claimed in claim 1 is characterized in that,
When moving above-mentioned data hidden, utilize grade point (removing the value of discrete cosine transform coefficient with the quantization parameter) service data in the piece of input picture operation quantization parameter and/or discrete cosine transform is hidden.
5, moving video encoding method as claimed in claim 1 is characterized in that,
Certain macro block synchronizing information to the piece group unit that forms said extracted when service data is hidden, is hidden in certain macro block synchronizing information of said extracted a piece group of the piece group back of the certain macro block synchronizing information of said extracted.
6, moving video encoding method as claimed in claim 1 is characterized in that,
Certain macro block synchronizing information that utilization is extracted from above-mentioned decoder, when run-time error was hidden, if the piece group macro block of input makes a mistake, just reference was from the macro block synchronizing information of the formation piece group of hiding data extraction, to the above-mentioned piece group macro block that makes a mistake, operation is encoded once more.
7, moving video encoding method as claimed in claim 6 is characterized in that,
When certain macro block synchronizing information of extracting from above-mentioned decoder belongs to the 6th the macro block synchronizing information that forms a piece group, if the above-mentioned macro block that makes a mistake is the macro block before the 6th macro block, just with reference to the synchronizing information of said extracted, from the 6th macro block of the above-mentioned piece group that makes a mistake, operation decoding once more.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410051733 CN1756350A (en) | 2004-09-29 | 2004-09-29 | Motion image coding method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410051733 CN1756350A (en) | 2004-09-29 | 2004-09-29 | Motion image coding method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1756350A true CN1756350A (en) | 2006-04-05 |
Family
ID=36689233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410051733 Pending CN1756350A (en) | 2004-09-29 | 2004-09-29 | Motion image coding method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1756350A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11277630B2 (en) | 2011-11-07 | 2022-03-15 | Dolby International Ab | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
| US11943485B2 (en) | 2011-11-07 | 2024-03-26 | Dolby International Ab | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
-
2004
- 2004-09-29 CN CN 200410051733 patent/CN1756350A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11277630B2 (en) | 2011-11-07 | 2022-03-15 | Dolby International Ab | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
| US11889098B2 (en) | 2011-11-07 | 2024-01-30 | Dolby International Ab | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
| US11943485B2 (en) | 2011-11-07 | 2024-03-26 | Dolby International Ab | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2944080B1 (en) | Method and apparatus for efficient coding of depth lookup table | |
| CN100348051C (en) | An enhanced in-frame predictive mode coding method | |
| CN1140133C (en) | Dual compressed video bitstream camera for universal serial bus connection | |
| RU2372743C2 (en) | Scalable video coding with two-level coding and one-level decoding | |
| CN109862356B (en) | Video coding method and system based on region of interest | |
| CN1361991A (en) | Video coding | |
| CN1757240A (en) | Video encoding | |
| CN1535018A (en) | Coding method of video frequency image jump over macro block | |
| AU2006230691A1 (en) | Video Source Coding with Decoder Side Information | |
| CN104065976B (en) | A kind of compression of images and security transmission method based on video | |
| CN1668110A (en) | Method and apparatus for improving the average image refresh rate in a compressed video bitstream | |
| CN101179719B (en) | Fast switch method of video sequence | |
| CN1196640A (en) | Method and decoder for decoding compressed moving-picture data | |
| CN1756350A (en) | Motion image coding method | |
| EP1484925A2 (en) | Method and device for compressing image data | |
| US20040228403A1 (en) | Moving picture coding method | |
| CN1633177A (en) | Method of frame rate adjusting for video communication system | |
| CN108833920A (en) | A kind of DVC side information fusion method based on light stream and Block- matching | |
| CN101223785A (en) | Encoding and decoding method and device for improving video error concealing | |
| US10887589B2 (en) | Block size determination for video coding systems and methods | |
| CN1926862A (en) | Video information transmission | |
| CN1219405C (en) | Image error hidding method | |
| US20040228404A1 (en) | Moving picture coding method | |
| CN102905129B (en) | Distributed coding method of still image | |
| CN104469377A (en) | H.264 file authentication playing method based on video steganography |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |