CN1633180A - Multi-description video coding method based on transformation and data fusion - Google Patents
Multi-description video coding method based on transformation and data fusion Download PDFInfo
- Publication number
- CN1633180A CN1633180A CN 200410075753 CN200410075753A CN1633180A CN 1633180 A CN1633180 A CN 1633180A CN 200410075753 CN200410075753 CN 200410075753 CN 200410075753 A CN200410075753 A CN 200410075753A CN 1633180 A CN1633180 A CN 1633180A
- Authority
- CN
- China
- Prior art keywords
- conversion
- description
- data fusion
- video coding
- coding method
- 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
A multi-description video encode method combining transformation and data includes applying transformation to being encoded signals and quantizing and entropy encoding the transformed signals to be decoded in terms of independent paths then to inversely transform the decoded signals to get the edge descriptions and merging the inversely transformed data to the central description, which can combines the multi-description encodes merging transformation and data with video codes. The encoding method can generate multiple MPEG code streams to one set of video sequence, and a video sequence of great distortion can be recovered from any stream, when multiple code streams are collected, a video sequence of small distortion will be recovered.
Description
Technical field
The invention belongs to technical field of video coding, more particularly relate to design based on the multi-description video coding method of conversion and data fusion.
Background technology
Multiple description coded (multiple description code) is encoded into two (or more than two) independently bit stream with individual signals, these independently bit stream be called descriptor (multiple descriptor).Multiple description coded have two main attributes: 1, can decode to each descriptor independently and reconstruct available primary signal: 2, there is complementary information in many descriptors, along with the increase of the descriptor quantity that correctly receives, the signal quality that decodes also progressively improves.It should be noted that first attribute is fundamentally different than conventional gradable (but or layering) coding, the basic layer that conventional method is encoded out is vital, if lose basic layer, other remaining bit stream is with utterly useless., multiple description coded technology can utilize any one descriptor that correctly receives to reconstruct useful primary signal, and along with the increase of the descriptor quantity that receives, the picture quality that decodes also progressively improves.Adopt the multi-description video coding algorithm, can also utilize in other descriptor unimpaired part to repair part impaired in this descriptor.Like this, even if two descriptors have all suffered packet loss, as long as the packet loss that these two descriptors suffer is not to take place simultaneously, they still can keep useful signal quality.Multiple description coded superiority also comprises: multiple description coded system can obtain higher compression efficiency, compares with conventional single descriptor (SD, single description) contraction principle, and the gross bit rate of its compression gained is only than latter Lve Gao; Multiple description coded system can support successful operation on the path of different bit rates; MPEG-4V2 (possessing the NEWPRED function), H.263V2 (possesses the RPS function) and H.264/AVC, the enhancement layer of these video encoding standards is all supported multi-description video coding.
Adopt multiple description coded technology, receive at least one descriptor, the primary signal of just can regenerating effectively as long as go up at any time.
Summary of the invention
Purpose of the present invention just is to overcome the shortcoming and defect of said method, and a kind of multi-description video coding method based on conversion and data fusion is provided.It combines multi-description video coding technology and multipath technology (multiple Path Diversity), transmits different descriptors with different network paths.Because the packet loss phenomenon in different paths is incoherent, therefore, MD coding techniques complimentary to one another between multi-path transmission system and the descriptor combines, and has just improved the efficient that the MD coding error restores greatly.
In order to achieve the above object, the present invention includes following steps:
1. to wanting encoded signals to implement conversion 1～n;
2. respectively the signal behind conversion 1～n is quantized and entropy coding;
3. respectively according to separately path 1～n to quantize and entropy coding after signal 1～n decode;
4. respectively decoded signal 1～n is carried out inverse transformation;
5. obtain the limit after the inverse transformation respectively and describe 1～n, the data fusion after 1～n the inverse transformation is become the center describe.
Based on conversion and data fusion multiple description coded is a kind of method in multiple description coded.Its basic thought is: implement different conversion to wanting encoded signals, the signal after the conversion is quantized and entropy coding, obtain a plurality of code streams.Each code stream all be one independently to the description of original signal, from any one code stream, can restore original signal.When a plurality of code streams are received, from each code stream, can restore a signal.These signals are got when all planting, can be restored a signal that quality is higher.
Wherein each description all is the code streams that meet the MPEG video standard, and the sequence of positions of the I frame described of each limit is just the same and be applied to above the I frame based on the multiple description coded of conversion and data fusion.
A conversion for each I frame is a dct transform, and other conversion is the displacement+dct transform to signal.Above-mentioned each code stream all meets MPEG1 standard or MPEG2 standard or mpeg 4 standard (comprising its second and the tenth part).
Video coding has many standards, and the codec of most of practical applications is all made according to standard.Concerning multiple description coded, when having only the descriptor standardization that produces in each code stream, just might be widely used.MPEG is present the most general method for video coding, and in the present invention, each code stream all will meet mpeg standard.
Describe for each, the order of I, P, B frame is just the same.The quantization error of having a mind to can be added to the limit of I frame and describe.The quantization error of having a mind to also can be added to the limit of P, B frame and describe.
The present invention is during to video signal coding, and the sequence of positions of I frame is duplicate in each is described.The position of I, P, B frame can be just the same under the situation of special case.
The present invention carries out multiple conversion to each I frame, and quantize with entropy coding to produce multiple description., in order to make each description can only be used dct transform by the decoding of the mpeg decoder of standard.In order to address this problem, in first is described, signal is used the dct transform of standard; In second description, can circulate clockwise to signal in the horizontal direction and move one.The reason of doing like this is that displacement+dct transform can be regarded as a kind of new conversion.After decoding, move one in decoding end as long as in the horizontal direction signal is circulated counterclockwise.Certainly, the direction of displacement and figure place also can adopt other value.Through after the conversion, signal is quantized and entropy coding.
In P, B frame, still carry out motion compensation, quantification and coding according to the method for stipulating in the standard.Because the different I frames of describing are different, thereby in different descriptions, the error of motion compensation also is different.In the process that quantizes, can add quantization error wittingly, the distortion factor of (central description) and limit description (side description) is described with the adjustment center.
The invention provides a kind of multi-description video coding method based on conversion and data fusion.It combines multi-description video coding technology and multipath technology (multiple Path Diversity), transmits different descriptors with different network paths.Because the packet loss phenomenon in different paths is incoherent, therefore, MD coding techniques complimentary to one another between multi-path transmission system and the descriptor combines, and has just improved the efficient that the MD coding error restores greatly.
The invention provides a kind of effectively video transmission coding method in real time on the internet.The Internet is a kind of network based on packet switch.With Circuit-switched network difference, the network of packet switch can make full use of bandwidth, but it is difficult to guarantee real-time.The present invention becomes the multichannel code stream with video signal compression, and different code streams can be from different path transmission.In most of the cases, have at least one road code stream in time to be received, thereby guarantee to obtain the signal of gross.When the multichannel code stream was received, higher-quality signal can be reduced out.
Description of drawings
Fig. 1 is multiple description coded and the structural representation multipath combination.
Fig. 2 is based on the multiple description coded structural representation of conversion and data fusion.
Fig. 3 is x[n when not adding quantization error],
With
The distributed architecture schematic diagram.
Fig. 4 is x[n when having a mind to add quantization error],
With
The distributed architecture schematic diagram.
Embodiment
Embodiment 1.A kind of multi-description video coding method based on conversion and data fusion, as shown in Figure 2.It may further comprise the steps:
1. to wanting encoded signals (1) to implement conversion 1～conversion n (2);
2. respectively the signal behind conversion 1～n (2) is quantized and entropy coding (3);
3. respectively according to separately path 1～n (4) to quantize and entropy coding (3) after signal 1～n decode (5);
4. respectively decoded signal 1～n is carried out inverse transformation (6);
5. obtain the limit after the inverse transformation respectively and describe 1～n, the data fusion (7) after 1～n the inverse transformation (6) is become the center describe.
Wherein each description all is the code streams that meet the MPEG video standard, and the sequence of positions of the I frame described of each limit is just the same and be applied to above the I frame based on the multiple description coded of conversion and data fusion.A conversion for each I frame is a dct transform, and other conversion is the displacement+dct transform to signal.Above-mentioned each code stream all meets MPEG1 standard or MPEG2 standard or mpeg 4 standard (comprising its second and the tenth part).MPEG is the image encoding compression standard that International Organization for Standardization passes through, and it is multiple to be divided into MPEG1, MPEG2, MPEG4 etc.Wherein MPEG1 passed through in 1992; MPEG2 is to the expansion of MPEG1 and improvement; MPEG4 presses the object tissue picture material, and has increased by 7 new functions.
Describe for each, the order of I, P, B frame is just the same.The quantization error of having a mind to can be added to the limit of I frame and describe.The quantization error of having a mind to also can be added to the limit of P, B frame and describe.
Fig. 1 shows multiple description coded and the structure multipath combination.Be encoded step (3) coding of signal among the figure (1) is decoded into by the decoded step in different path (4) (5) then that the limit is described and the center description.
Fig. 3 and shown in Figure 4, x[n] be the value of signal on the n of position,
Be the value of description one on the n of position,
Be the value of description two on the n of position.According to above-mentioned method, in the center is described, the value on the n of position by
With
Evenly get.Though Fig. 3 does not add x[n when changing error],
With
Distribution.If have a mind to add quantization error, then as shown in Figure 4, the error of center description will reduce so.Certainly, do like this to strengthen and describe two error, whether add this quantization error and will describe and describe one importance on the center and decide.
Claims (9)
1. multi-description video coding method based on conversion and data fusion, 1. it comprise wanting step that encoded signals implements conversion 1～n 2. respectively the signal behind conversion 1～n to be quantized step with entropy coding, it is characterized in that further comprising the steps of:
3. respectively according to separately path 1～n to quantize and entropy coding after signal 1～n decode;
4. respectively decoded signal 1～n is carried out inverse transformation;
5. obtain the limit after the inverse transformation respectively and describe 1～n, the data fusion after 1～n inverse transformation becomes the center and describes.
2. according to the described multi-description video coding method of claim 1 based on conversion and data fusion, it is characterized in that wherein each description all is the code streams that meet the MPEG video standard, and the sequence of positions of the I frame described of each limit is just the same and be applied to above the I frame based on the multiple description coded of conversion and data fusion.
3. according to the described multi-description video coding method based on conversion and data fusion of claim 2, it is characterized in that a conversion for each I frame is a dct transform, other conversion is the displacement+dct transform to signal.
4. according to the described multi-description video coding method of claim 2, it is characterized in that said each code stream all meets the MPEG1 standard based on conversion and data fusion.
5. according to the described multi-description video coding method of claim 2, it is characterized in that said each code stream all meets the MPEG2 standard based on conversion and data fusion.
6. according to the described multi-description video coding method of claim 2, it is characterized in that said each code stream all meets mpeg 4 standard based on conversion and data fusion.
7. according to the described multi-description video coding method of claim 2, it is characterized in that the order of I, P, B frame is just the same for each is described based on conversion and data fusion.
8. according to the described multi-description video coding method of claim 2, it is characterized in that the quantization error of having a mind to is added in the limit description of I frame based on conversion and data fusion.
9. according to the described multi-description video coding method of claim 2, it is characterized in that the quantization error of having a mind to is added in the limit description of P, B frame based on conversion and data fusion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410075753 CN1633180A (en) | 2004-12-24 | 2004-12-24 | Multi-description video coding method based on transformation and data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410075753 CN1633180A (en) | 2004-12-24 | 2004-12-24 | Multi-description video coding method based on transformation and data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1633180A true CN1633180A (en) | 2005-06-29 |
Family
ID=34846962
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410075753 Pending CN1633180A (en) | 2004-12-24 | 2004-12-24 | Multi-description video coding method based on transformation and data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1633180A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100391264C (en) * | 2005-09-25 | 2008-05-28 | 海信集团有限公司 | Multi-descriptive coding-decoding method based on AVS video stardard |
| WO2009056071A1 (en) * | 2007-10-26 | 2009-05-07 | Huawei Technologies Co., Ltd. | A multiple description coding and decoding method, system and apparatus based on frame |
| CN101394569B (en) * | 2008-10-29 | 2010-04-21 | 北京创毅视讯科技有限公司 | Error tolerance method, device and chip for AVS video decoder |
| CN102523448A (en) * | 2011-12-27 | 2012-06-27 | 北京交通大学 | Multiple description video coding and decoding method based on rotation, apparatus and system thereof |
| CN103081473A (en) * | 2010-07-31 | 2013-05-01 | 吴秀美 | Prediction block generating device |
| CN101809873B (en) * | 2007-10-05 | 2013-11-06 | 爱立信电话股份有限公司 | Method and apparatus for multiple description coding |
-
2004
- 2004-12-24 CN CN 200410075753 patent/CN1633180A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100391264C (en) * | 2005-09-25 | 2008-05-28 | 海信集团有限公司 | Multi-descriptive coding-decoding method based on AVS video stardard |
| CN101809873B (en) * | 2007-10-05 | 2013-11-06 | 爱立信电话股份有限公司 | Method and apparatus for multiple description coding |
| WO2009056071A1 (en) * | 2007-10-26 | 2009-05-07 | Huawei Technologies Co., Ltd. | A multiple description coding and decoding method, system and apparatus based on frame |
| CN101420607B (en) * | 2007-10-26 | 2010-11-10 | 华为技术有限公司 | Method and apparatus for multi-description encoding and decoding based on frame |
| CN101394569B (en) * | 2008-10-29 | 2010-04-21 | 北京创毅视讯科技有限公司 | Error tolerance method, device and chip for AVS video decoder |
| CN103081473A (en) * | 2010-07-31 | 2013-05-01 | 吴秀美 | Prediction block generating device |
| CN102523448A (en) * | 2011-12-27 | 2012-06-27 | 北京交通大学 | Multiple description video coding and decoding method based on rotation, apparatus and system thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1192629C (en) | System and method for improved fine granular scalable video using base layer coding information | |
| JP4922391B2 (en) | Multi-layer video encoding method and apparatus | |
| US8130830B2 (en) | Enhancement layer switching for scalable video coding | |
| CN1253014C (en) | Image coding method and apparatus an image decoding method and apparatus | |
| US20020009233A1 (en) | Color encoding and decoding method | |
| CN1575606A (en) | Spatial scalable compression | |
| JP2003523667A (en) | System and method using pre-predictive bitplane coding for progressive fine granularity scalable (PFGS) video coding | |
| US20060159359A1 (en) | Fine granularity scalable video encoding and decoding method and apparatus capable of controlling deblocking | |
| CN1526240A (en) | Method and device for generating a scalable coded video signal from a non-scalable coded video signal | |
| JP2005507590A5 (en) | ||
| CN1229324A (en) | Method and apparatus for encoding video signal | |
| WO2007069829A1 (en) | Method and apparatus for encoding and decoding video signals on group basis | |
| US20060008002A1 (en) | Scalable video encoding | |
| CN1672421A (en) | Method and apparatus for performing multiple description motion compensation using hybrid predictive codes | |
| CN1633180A (en) | Multi-description video coding method based on transformation and data fusion | |
| CN1656816A (en) | Improved efficiency fgst framework employing higher quality reference frames | |
| JP3576660B2 (en) | Image encoding device and image decoding device | |
| CN100499815C (en) | Video frequency coding and de-coding method for supporting video frequency frame random reading | |
| CN1202673C (en) | Enhanced type fineness extensible video coding structure | |
| CN100391264C (en) | Multi-descriptive coding-decoding method based on AVS video stardard | |
| JP3914214B2 (en) | Image coding apparatus and image decoding apparatus | |
| AU2012100851A4 (en) | Multiple description video codec based on adaptive temporal sub-sampling | |
| Mansour et al. | Low redundancy layered multiple description scalable coding using the subband extension of H. 264/AVC | |
| Liu et al. | Low-delay distributed multiple description coding for error-resilient video transmission | |
| Yu et al. | Image recombination based side information reconstruction in Wyner-Ziv video coding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20050629 |