CN201282535Y - Device for converting H.264 to AVS video code stream - Google Patents
Device for converting H.264 to AVS video code stream Download PDFInfo
- Publication number
- CN201282535Y CN201282535Y CN 200820146470 CN200820146470U CN201282535Y CN 201282535 Y CN201282535 Y CN 201282535Y CN 200820146470 CN200820146470 CN 200820146470 CN 200820146470 U CN200820146470 U CN 200820146470U CN 201282535 Y CN201282535 Y CN 201282535Y
- Authority
- CN
- China
- Prior art keywords
- frame
- multichannel
- transcoder
- avs
- domain
- 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 - Lifetime
Links
Images
Abstract
The utility model discloses a device for converting a video code stream from H.264 to AVS, and is characterized in that the utility model comprises an H.264 decoder, a rapid cascade transcoder based on a pixel domain, an I-frame compressed domain transcoder, an I-frame sequence output module, a P-frame intra-frame compressed domain transcoder, a P-frame inter-frame compressed domain transcoder, a P-frame sequence output module, a B-frame intra-frame compressed domain transcoder, a B-frame inter-frame compressed domain transcoder, a B-frame sequence output module , a whole rapid transcoder, and an AVS code stream output module. Based on the rapid cascade transcoding strategy from H.264 to AVS, an image information reuse mode, a QP mapping method, and a motion vector mapping based on the pixel domain, the utility model avoids the complex computation with great arithmetic and meets the application requirement of high-definition videos to be transcoded on a real-time basis.
Description
Technical field
The utility model belongs to multimedia communication and video transmission technologies field, particularly relates to H.264 form to the device of the video format conversion between the AVS form.
Background technology
H.264 be the video encoding standard of new generation of joint video team (JVT:JOINT VIDEO TEAM) exploitation of the MPEG (moving picture expert group) of the VCEG (video coding expert group) of ITU-T and ISO/IEC, it be ITU-T H.264, be again the 10th part of the MPEG-4 of ISO/IEC.Begin to collect draft in 1998, formally became international standard in 2003.H.264 adopted multinomial new technology in the middle of, as unified VLC symbolic coding, high accuracy, multimodal Displacement Estimation are based on 4 * 4 integer transforms, layered encoded grammer etc.These measures make that H.264 algorithm has very high code efficiency, under identical reconstructed image quality, with respect to H.263 or the MPEG-2 video compression than improving 1 times, or save 50% code check.Defined VLC and NAL layer respectively, therefore to network, particularly IP and wireless network have good compatibility.H.264 defined the video coding of basic class, main file time and three class of expansion class, each class is supported specific encoding function, and has stipulated the requirement to corresponding codec.
Because historical reasons, coding and decoding video algorithm standard is all grasped in foreign vendor's hand of minority all the time, as adopting maximum H.264 coding and decoding video algorithms in the present digital remote viewing system, grasped by the patent manufacturer of representative with American-European, Japan and Korea S exactly, these patent manufacturers are by collecting the purpose that high patent fee is used for reaching the Chinese video technique development of restriction.
For promoting the development of Chinese audio frequency and video national industry, set up Chinese digital audio/video encoding and decoding technique standard operation group in 2002 in Beijing, be called for short AVS, its working range is the information industry demand towards China, associating domestic enterprise and scientific research institution, system (repairing) is ordered the compression of digital audio/video, decompress, common technology standards such as processing and expression, for digital audio/video equipment and system provide the encoding and decoding technique of high-efficiency and economic, serve high-resolution digital broadcasting, the high-density laser digital storage media, the WiMAX multi-media communication, great information industry such as the Internet broadband Streaming Media is used.
AVS is the China national source encoding standard of formulating " digital audio/video encoding and decoding technique standard operation group " (being called for short AVS working group), at present by international endorsement.AVS because of at the applied customization standard, has overcome H.264 redundant shortcoming simultaneously H.264 to be starting point, has reached well compromise between complexity and code efficiency.AVS is a cover adaptive surface very wide technical standard, and advantage shows the following aspects: (1) AVS be based on China's autonomous innovation technology and international public technology constructed from primary standard, properly settled Intellectual Property Rights Issues; (2) code efficiency height is than the high 2-3 of MPEG-2 international standard times (high definition TV can reach 3 times or more); (3) computation complexity is low, and hardware implementation cost is lower; (4) AVS can save wireless frequency spectrum over half and wire message way resource, significantly reduces the economy input of transmission, memory device and system.AVS has adopted H.264 similar techniques technically.Code efficiency is 2-3 times (different according to video pictures size difference) of MPEG-2 video, surpassed international MPEG-4 AVC/H.264 standard, and scheme is succinct, and implementation complexity is lower than MPEG-4 AVC.
The basic ideas of China AVS working group are at layer employing and up-to-date international standard ISO MPEG-4/AVC and H.264 similar techniques framework substantially, help chip and support the world and Chinese Industrial Standards (CIS) simultaneously; Adopt proprietary technology at MP (Main Profile) layer, effectively avoid of the patent charge of international manufacturer the MP layer; On technical scheme, simplify some unnecessary modules that add because of non-technical reason in the international standard, thereby make the cost performance of this technology higher.In March, 2006, official approval the national standard of AVS-P2 as video coding and decoding technology.For evading following MPEG-4 and the H.264 huge patent fee puzzlement of standard, the inner appearance policy of the Ministry of Information Industry and the Department of Science and Technology advises that audiovisual applications adopts homemade AVS standard.
Along with the promotion and implementation of AVS, new video-unit will adopt AVS video national standard to compress and transmit.And existing video-unit major part is based on video compression standard H.264, in order to save cost, fully to reuse, produced from form H.264 to the requirement of AVS form type code stream translation.Provide high-quality H.264 form to the quick conversion plan of AVS format video, become the important technological problems that the present technique field needs to be resolved hurrily.
The utility model content
For solving the problems of the technologies described above, avoid the complex calculation of intensive in the transcoding algorithm, the device that the image laser propagation effect that provide high-quality, easily realize, real-time is good, the utility model provide form H.264 to change to AVS format video code stream.
A kind of AVS video code flow conversion equipment that H.264 arrives comprises:
H.264 decoder is used for that code stream is H.264 carried out decompression operation and obtains the intermediate code flow data;
Based on the quick cascade transcoder of pixel domain, be used for the high level syntax structure of middle bit stream data is carried out handling based on the quick cascade transcoding of pixel domain the high level syntax structure of output AVS code stream;
I frame compression domain transcoder is used for the I frame of middle bit stream data is carried out the processing of compression domain transcoding, and exports the I frame sequence of AVS code stream by I frame sequence output module;
Compression domain transcoder in the P frame frame, being used for that the macro block that adopts intraframe coding in the middle bit stream data P frame is carried out the compression domain transcoding handles, P frame interframe compression domain transcoder, be used for that the macro block that adopts interframe encode in the middle bit stream data P frame is carried out the compression domain transcoding and handle, and export the P frame sequence of AVS code stream by P frame sequence output module;
Compression domain transcoder in the B frame frame, being used for that the macro block that adopts intraframe coding in the middle bit stream data B frame is carried out the compression domain transcoding handles, B frame interframe compression domain transcoder, be used for that the macro block that adopts interframe encode in the middle bit stream data B frame is carried out the compression domain transcoding and handle, and export the B frame sequence of AVS code stream by B frame sequence output module;
Whole fast transcoding device is used for that the bit stream data after the above-mentioned conversion is carried out whole fast transcoding and handles, at last by the AVS code stream after the AVS code stream output module output conversion.
A kind of AVS video code flow conversion equipment that H.264 arrives comprises:
H.264 multi-channel decoder, be used for to multichannel H.264 code stream carry out the parallel decompression operation and obtain multichannel intermediate code flow data;
Based on the quick cascade multichannel transcoder of pixel domain, be used for multichannel intermediate code flow data high level syntax structure is carried out handling based on the parallel quick cascade transcoding of pixel domain the high level syntax structure of output multi-channel A VS code stream;
I frame compression domain multichannel transcoder is used for the I frame of multichannel intermediate code flow data is carried out parallelly compressed domain code conversion processing, and exports the I frame sequence of AVS code stream by I frame sequence multichannel output module;
Compression domain multichannel transcoder is used for the macro block that adopts intraframe coding in the multichannel intermediate code flow data P frame is carried out parallelly compressed domain code conversion processing in the P frame frame, P frame interframe compression domain multichannel transcoder is carried out parallelly compressed domain code conversion processing to the macro block that adopts interframe encode in the P frame in the multichannel intermediate code flow data, and exports the P frame sequence of AVS multichannel code stream by P frame sequence multichannel output module;
Compression domain multichannel transcoder is used for the macro block that adopts intraframe coding in the multichannel intermediate code flow data B frame is carried out parallelly compressed domain code conversion processing in the B frame frame, B frame interframe compression domain multichannel transcoder is carried out parallelly compressed domain code conversion processing to the macro block that adopts interframe encode in the B frame in the multichannel intermediate code flow data, and exports the B frame sequence of AVS multichannel code stream by B frame sequence multichannel output module;
Whole multichannel transcoder fast is used for the whole fast transcoding that walk abreast of the multichannel bit stream data after the above-mentioned conversion is handled, and exports multi-channel A VS code stream by multi-channel A VS output module at last.
As from the foregoing, described in the utility model be H.264 form to the device of AVS format video code stream conversion, reuse based on quick cascade transcoding architecture, the motion vector of pixel domain by adopting, the motion residual error is reused, the DCT/IDCT sequence is reset, the video sequence header is reused, macro-block coding pattern information such as reuses at key technology, complex calculations have been avoided, made full use of the H.264 image and the structural information of source code flow, thereby realized that H.264 form is to the quick conversion of AVS format video code stream.
Description of drawings
Fig. 1 is the quick cascade transcoder structure chart based on pixel domain that the utility model adopts;
Fig. 2 is that H.264 the utility model arrives AVS video code flow conversion equipment workflow diagram;
Fig. 3 is that H.264 the utility model arrives AVS video code flow conversion device structure schematic diagram.
Fig. 4 is that H.264 the utility model arrives AVS video code flow conversion equipment multichannel processing structural representation
Embodiment
Describe the technical solution of the utility model in detail below in conjunction with specific embodiment of the utility model.
H.264 all be based on the conversion and the prediction hybrid coding algorithm of piece with AVS, the basic structure of two kinds of encoders is similar, and its technical characteristic contrast sees Table 1.
Table 1
| Technology modules | AVS-P2 | H.264 |
| Infra-frame prediction | Based on 8 * 8,5 kinds of luma prediction modes, 4 kinds of prediction mode for chroma | 9 kind of 4 * 4 luminance block pattern, 4 kind of 16 * 16 luminance block pattern, 4 kind of 8 * 8 chrominance block pattern |
| The multi-reference frame prediction | Maximum 2 frames | The buffer management mechanism of maximum 16 frame complexity |
| Become the block-sized motion compensation | 16 * 16,16 * 8,8 * 16,8 * 8 motion search | 16 * 16,16 * 8,8*16,8*8,8*4,4 * 8,4 * 4 motion search |
| B frame macro block symmetric pattern | Only searching for forward motion vector gets final product | Bidirectional research |
| The compensation of 1/4 pixel motion | 1/2 location of pixels adopts 4 to clap filtering 1/4 location of pixels employing 4 bat filtering, linear interpolation | 1/2 location of pixels adopts 6 to clap filtering 1/4 location of pixels linear interpolation |
| Conversion and quantification | Decoding end normalization is finished at coding side, reduces decoding complexity | The encoding and decoding end all need carry out normalization |
| Entropy coding | Context-adaptive 2D-VLC, the Exp-Golomb sign indicating number reduces calculating and memory complexity | CAVLC: with piece correlation height on every side, realize more complicated CABAC: hardware is realized complicated especially |
| Loop filtering | Carry out simple filtering strength classification, the pixel that filtering is less based on 8 * 8 block edges | Carry out based on 4 * 4 block edges, the filtering strength classification is various, and filter edge is many |
| Fault-tolerant encoding | The simple band mechanism of dividing is enough to satisfy error concealing, the recovery demand in the extensive application | Data are cut apart, complicated macro blocks such as FMO/ASO, band organization mechanism, pressure Intra piece refreshes coding, restrictive infra-frame prediction etc. and realize complicated especially |
H.264 technical characterictic: support weight estimation; Support arbitrary resolution image encoding (QCIF～D1); Macro block (mb) type and all predictive modes such as intraframe coding support 16 * 16,8 * 8,4 * 4; Interframe encode is supported 16 * 16 to 4 * 4 various macroblock partitions methods; The constant quantization parameter is supported in the Rate Control aspect, supports single channel and multichannel ABR; Support self-defined quantization table; Support the multi-ribbon parallel encoding.Do not support interleaved image; Do not support PAFF and MBAFF that field mode is relevant.
The technical characterictic of AVS: support arbitrary resolution image encoding (QCIF～SD/D1); 5 kind of 8 * 8 luminance block pattern and 4 kind of 8 * 8 chrominance block pattern are supported in intraframe coding; Interframe encode is supported 16 * 16,16 * 8,8 * 16 and 8 * 8 four kinds of macroblock partitions types; Support 3 kinds of B frame prediction modes; Maximum 2 frames of B frame number; Support RDO; Support FME; Do not support interleaved image.
The basic processing unit of video decoding process is a macro block among the AVS, and a macro block comprises one 16 * 16 luminance block and corresponding chrominance block.Macro block can further be divided into the piece of minimum 8 * 8 and predict.The unit that changes is 8 * 8 piece.Conversion coefficient carries out scalar quantization.In addition, the AVS processing method of transform and quantization in the restricted code device not.
In the motion prediction of AVS, a macro block can be divided into different sub-pieces according to Fig. 2, forms the piece size of 4 kinds of different modes.In this manner, in each macro block, include 1,2 or 4 motion vector.Motion vector and the order of reference key in code stream after the numeral macroblock partitions among Fig. 2 in the rectangle.
The H.264 form that the utility model provides is to AVS format video code stream conversion method, and the spatial resolution and the frame rate of video do not change before and after transcoding, and just coding standard change has taken place and data transmission rate changes thereupon.Greatly reduce the complexity of translation operation.
The converting video code stream that the utility model mainly adopts is the compression domain transcoding, promptly only utilize in the H.264 code stream of input information compressed, as the DCT coefficient after video sequence header, macro-block coding pattern information, motion vector information and the quantification etc., directly generate the AVS code stream after the conversion.
The H.264 quick cascade transcoder structure based on pixel domain in the AVS video code flow conversion equipment that the utility model provides is as follows, (1) reuses motion vector in the input code flow H.264, to its module of refining and only needing search in the subrange and not carrying out the global motion estimation again; (2) by to obtain the module of the prediction residual of AVS do motion compensation from the prediction residual in the code stream H.264; (3) reduce the module of the operation of some DCT/IDCT by the sequence of motion that rearranges DCT/IDCT.
H.264, what the utility model provided carries out quick cascade transcoder (as Fig. 1) based on pixel domain to code stream H.264 in the AVS video code flow conversion equipment, the high level syntax structure of AVS video code flow is constant behind the maintenance transcoding, it is the invariant position of sequence head, invariant position and the structure of GOP are constant, the type of prediction of frame is constant, be still to be the I frame behind the I frame transcoding, still the P frame behind the P frame transcoding, still the B frame behind the B frame transcoding, keep the scan type of frame constant, and repartition slice (be defaulted as each frame and only be divided into 1 slice).Adopt this mode except amount of calculation is little, also avoided the frame reorder delay.
The H.264 I frame compression domain transcoder in the device of AVS video code flow conversion that the utility model provides is carried out the compression domain transcoding to the I frame in the input code flow H.264, recode behind elder generation's partial decoding of h for the AVS code stream and by the output of I frame sequence output module, each macro block of I frame is all adopted the intraframe coding algorithm recompile of AVS, when coding with reference to corresponding macroblock encoding information in the input code flow H.264, reusing H.264, inlet flow quantizes parameter.
In the H.264 P frame frame in the device of AVS video code flow conversion that the utility model provides in transcoder and the B frame frame transcoder respectively to adopting the compression domain transcoding of the macro block of intraframe coding in P frame and the B frame, again carrying out macro-block coding pattern during transcoding selects and motion compensation, reuse the motion vector in the inlet flow and do not carry out the global motion estimation again, by to obtaining AVS motion residual error behind the transcoding from the prediction residual in the inlet flow H.264 do motion compensation, reusing H.264, inlet flow quantizes parameter.
H.264 P frame interframe transcoder in the device of AVS video code flow conversion that the utility model provides and B frame interframe transcoder are respectively to adopting the compression domain transcoding of the macro block of interframe encode in P frame and the B frame, prediction piece size is handled, carried out fritter for the unsupported macro block rank of AVS and merge.Reference frame is shone upon.For the two-way B frame prediction in H.264, the forward motion vector in only keeping H.264, and be mapped to symmetric pattern among the AVS.Reusing H.264, inlet flow quantizes parameter.
The utility model provide H.264 in the device of AVS video code flow conversion inter prediction mainly comprise with the lower part:
A: the module of processed pixels accuracy prediction, H.264 all support not process the motion vector of 1/4 pixel (brightness) and 1/8 pixel (colourity) precision in precision with AVS-P2.
B: the module of handling the block size prediction, because the branch block mode in H.264 has 7 kinds, and has only 4 kinds in AVS-P2, so done following processing, (1) to the four kinds of branch block modes of 16 * 16,16 * 8,8 * 16 and 8*8 in H.264, the MV that obtains in can directly utilizing H.264.For asking further accurately, with the position of this MV prediction up and down 4 points search for, be that matching criterior finds optimum with SAD again; (2) for the unsupported macro block rank of AVS, carry out fritter and merge, carry out tentative prediction according to the MV before the transcoding again, find match point preferably near the less hunting zone this predicted position again.
C: the selection module of handling reference frame, H.264 the reference frame that uses in inter prediction is more, and the B frame also can be used as the reference frame use, and use the decoded two continuous frames in its front for referencial use to the P frame among the AVS-P2, use before and after it each frame as a reference to the B frame, so let us do a reference frame mapping when designing on reference frame.
D: the module of treatments B frame prediction, to the two-way B frame prediction in H.264, support forward motion vector in the symmetric pattern because have only among the AVS-P2, thus the forward motion vector in only having kept H.264, and be mapped to symmetric pattern among the AVS-P2.
The utility model provide H.264 in the device of AVS video code flow conversion for inlet flow H.264 quantize parameter (QP) to reuse part as follows:
Before the quantization parameter in reusing H.264, H.264 quantization parameter is mapped on the quantization parameter of AVS, could makes like this can guarantee the minimal data distortion in theory in the transcoding process.Because H.264 the quantization principles with AVS is identical (quantization step all is an exponential distribution), and QP scope also identical (0～63), so this mapping is finished than being easier to.In AVS-P2, go for this quantization step corresponding quantitative parameter with H.264 quantization step,, then get its corresponding quantitative parameter, finish mapping with the absolute value minimum of the difference of other quantization step if do not have identically.Table look-up the time for minimizing, utilize among the AVS-P2, the every increase by 8 of QP, quantization step increases by 1 times characteristic, earlier be that quantization parameter corresponding quantitative step-length is searched for by unit, find the absolute value minimum of difference, and then in these 8 quantization steps, make further search refinement with 8.
QP mapping content in the utility model is as follows:
The AVS quantization step computing formula that adopts:
Qstepavs≈(2(QPavs//8))*QP2QSTEPavs[QPavs％8] (1)
The H.264/AVC computing formula that adopts:
Qstep264≈(2(QP264//6)*QP2QSTEP264[QP264％6] (2)
All AVS and H.264/AVC step-length can be according to formula (1), formula (2) is calculated.For with same ratio from H.264/AVC being transcoded onto AVS, we can select QP to make AVS and the difference of Qstep H.264 minimum.
Mode mapping content in the utility model is as follows:
Because the resolution of SD or HD video is very big comparatively speaking, remove 8*4,4*8, the inter-frame forecast mode of 4*4 piece can reduce complexity with less loss, if QStepavs ≈ Qstep264, so AVS and H.264 between quantization error about the same, so Ravs ≈ R264 is rational in code check distortion theory, although always the same in reality, experimental result shows that the caused code efficiency loss of re-use pattern therefor can be left in the basket.
The motion vector mapping content that the utility model comprises is as follows:
The integer pixel motion vector can be reused, and the motion vector of 1/2 and 1/4 pixel also can be reused so.It is as follows that AVS is used to calculate the formula of 1/2 location of pixels:
Favs＝(-1，5，5，-1)/8 (3)
Corresponding H.264 be:
F264＝(1，-5，20，20，-5，1)/32 (4)
By formula (3) and formula (4), we can directly calculate AVS and H.264/AVC between the difference of pixel value.For example:
bavs＝(-F+5*G+5*H-I)/8 (5)
H.264/AVC half-pix is:
b264＝(E-5*F+20*G+20*H-5*I+J)/32 (6)
Both differences are:
θ＝bavs-b264＝(-E+F+I-J)/32 (7)
Clearly, if-E+F+I-J is less than 32, and θ will equal 0, and when image especially smoothly is, this situation is to exist, and the worst situation causes differing from 16, but this situation often do not take place, and institute will can not lose much according to coding efficiency.
Similarly, the situation that we can reasoning 1/4 pixel also is the same.
By re-use pattern therefor and motion vector mapping that the utility model adopts, H.264 the complexity to AVS video conversion Calculation can greatly reduce.
Be illustrated in figure 2 as H.264 form to the overall workflow of AVS format video code stream conversion equipment.
As shown in Figure 3, H.264 form comprises in decoder H.264, the quick cascade transcoder, I frame compression domain transcoder, I frame sequence output module, P frame frame based on pixel domain compression domain transcoder, B frame interframe compression domain transcoder, B frame sequence output module, whole fast transcoding device, AVS code stream output module in compression domain transcoder, P frame interframe compression domain transcoder, P frame sequence output module, the B frame frame to AVS format video code stream conversion equipment.Wherein:
H.264 decoder, be used for H.264 input code flow decoding, and produce H.264 output code flow, decoding meets the decoding standard in the decoder H.264 in decoder H.264, and detail can be with reference to the 8th part among the ITU-T Rec.H.264|ISO/IEC International Standard11496-10.Advanced Video Coding for generic audiovisual services.2005 (E).
Be used for bit stream data high level syntax structure is H.264 carried out handling based on the quick cascade transcoding of pixel domain based on the quick cascade transcoder of pixel domain, obtain the high level syntax structural framing of AVS output code flow;
I frame compression domain transcoder is handled in the I frame in the bit stream data is H.264 carried out the compression domain transcoding, and bit stream data is obtained the I frame sequence of AVS output code flow after by the output of I frame sequence output module;
The compression domain transcoder is used for H.264 adopting the macro block of intraframe coding to carry out the processing of compression domain transcoding in the bit stream data P frame in the P frame frame, P frame interframe compression domain transcoder is used for that the macro block that adopts interframe encode in the bit stream data P frame is H.264 carried out the compression domain transcoding to be handled, and by obtaining the P frame sequence of AVS output code flow after the output of P frame sequence output module;
Transcoder is used for H.264 adopting the macro block of intraframe coding to carry out the processing of compression domain transcoding in the bit stream data B frame in the B frame frame, B frame interframe compression domain transcoder is used for that the macro block that adopts interframe encode in the bit stream data B frame is H.264 carried out the compression domain transcoding to be handled, and the B frame sequence by obtaining the AVS output code flow after the output of B frame sequence output module;
Whole fast transcoding device is used for that bit stream data is H.264 carried out whole fast transcoding to be handled, and by obtaining the AVS output code flow after the output of AVS code stream output module.
A kind of AVS video code flow conversion equipment that H.264 arrives that the utility model proposes, as shown in Figure 4, comprise H.264 in multi-channel decoder, quick cascade multichannel transcoder, I frame compression domain multichannel transcoder, I frame sequence multichannel output module, the P frame frame compression domain multichannel transcoder, B frame interframe compression domain multichannel transcoder, B frame sequence multichannel output module, whole multichannel transcoder, AVS code stream multichannel output module fast in compression domain multichannel transcoder, P frame interframe compression domain multichannel transcoder, P frame sequence multichannel output module, the B frame frame based on pixel domain:
H.264 multi-channel decoder be used for to multichannel H.264 code stream carry out parallel decompression operation and obtain H.264 bit stream data of multichannel;
Based on the quick cascade multichannel transcoder of pixel domain be used for to multichannel H.264 bit stream data high level syntax structure carry out handling based on the parallel quick cascade transcoding of pixel domain, obtain the high level syntax structural framing of multi-channel A VS output code flow, and the high level syntax structure of multi-channel A VS video code flow is constant behind the maintenance transcoding, it is the invariant position of sequence head, invariant position and the structure of GOP are constant, the type of prediction of frame is constant, be still to be the I frame behind the I frame transcoding, still the P frame behind the P frame transcoding, still the B frame behind the B frame transcoding, keep the scan type of frame constant, and repartition slice (be defaulted as each frame and only be divided into 1 slice).Adopt this mode except amount of calculation is little, also avoided the frame reorder delay.；
I frame compression domain multichannel transcoder be used for to multichannel H.264 the I frame of bit stream data carry out parallelly compressed domain code conversion and handle, and the I frame sequence by I frame sequence multichannel output module output multi-channel A VS code stream;
Compression domain multichannel transcoder is used for H.264 adopting the macro block of intraframe coding to carry out parallelly compressed domain code conversion processing in the bit stream data P frame to multichannel in the P frame frame, P frame interframe compression domain multichannel transcoder is used for H.264 adopting the macro block of interframe encode to carry out parallelly compressed domain code conversion processing in the bit stream data P frame to multichannel, and exports the P frame sequence of multi-channel A VS code stream by P frame sequence multichannel output module;
Compression domain multichannel transcoder is used for H.264 adopting the macro block of intraframe coding to carry out parallelly compressed domain code conversion processing in the bit stream data B frame to multichannel in the B frame frame, B frame interframe compression domain multichannel transcoder is used for H.264 adopting the macro block of interframe encode to carry out parallelly compressed domain code conversion processing in the bit stream data B frame to multichannel, and exports the B frame sequence of multi-channel A VS bit stream data by B frame sequence multichannel output module;
Whole multichannel transcoder fast be used for to multichannel H.264 the bit stream data whole fast transcoding that walk abreast handle, and export multi-channel A VS bit stream data by AVS code stream multichannel output module.
In sum, H.264, the utility model provides arrives AVS video code flow conversion equipment, make full use of H.264 feature with the AVS structural similarity, make full use of the H.264 image and the structural information of source code flow, employing is based on fast transcoding system, image information re-use pattern therefor, QP mapping and the motion vector mapping of pixel domain, avoid the complicated calculations and the conversion of macrooperation amount, realized H.264 arriving the quick conversion of AVS video code flow.
Obviously, those skilled in the art can carry out various changes and modification to the embodiment in the utility model and not break away from spirit and scope of the present utility model.Like this, if these among the utility model embodiment are revised and modification belongs within the scope of the utility model claim and equivalent technologies thereof, then the embodiment in the utility model also is intended to comprise these changes and modification interior.
Claims (2)
1, a kind of AVS video code flow conversion equipment that H.264 arrives is characterized in that, comprising:
H.264 code stream is carried out the H.264 decoder that decompression operation obtains the intermediate code flow data;
The high level syntax structure of middle bit stream data is carried out handling based on the quick cascade transcoding of pixel domain the quick cascade transcoder based on pixel domain of the high level syntax structure of output AVS code stream;
I frame in the middle bit stream data is carried out the compression domain transcoding handle, and export the I frame compression domain transcoder of the I frame sequence of AVS code stream by I frame sequence output module;
The macro block that adopts intraframe coding in the P frame in the middle bit stream data is carried out compression domain transcoder in the P frame frame that the compression domain transcoding handles;
The macro block that adopts interframe encode in the P frame in the middle bit stream data is carried out the compression domain transcoding handle, and export the P frame interframe compression domain transcoder of the P frame sequence of AVS code stream by P frame sequence output module;
The macro block that adopts intraframe coding in the B frame in the middle bit stream data is carried out compression domain transcoder in the B frame frame that the compression domain transcoding handles;
The macro block that adopts interframe encode in the B frame in the middle bit stream data is carried out the compression domain transcoding handle, and export the B frame interframe compression domain transcoder of the B frame sequence of AVS code stream by B frame sequence output module;
Bit stream data after the above-mentioned transcoder conversion is carried out whole fast transcoding handle, the whole fast transcoding device of the AVS code stream after changing by the output of AVS code stream output module at last.
2, a kind of AVS video code flow conversion equipment that H.264 arrives is characterized in that, comprising:
To multichannel H.264 code stream carry out the H.264 multi-channel decoder that parallel decompression operation obtains multichannel intermediate code flow data;
Multichannel intermediate code flow data high level syntax structure is carried out handling based on the parallel quick cascade transcoding of pixel domain the quick cascade multichannel transcoder based on pixel domain of the high level syntax structure of output multi-channel A VS code stream;
I frame in the multichannel intermediate code flow data is carried out parallelly compressed domain code conversion handle, and export the I frame compression domain multichannel transcoder of the I frame sequence of AVS code stream by I frame sequence multichannel output module;
The macro block that adopts intraframe coding in the P frame in the multichannel intermediate code flow data is carried out compression domain multichannel transcoder in the P frame frame that parallelly compressed domain code conversion handles;
The macro block that adopts interframe encode in the P frame in the multichannel intermediate code flow data is carried out parallelly compressed domain code conversion handle, and export the P frame interframe compression domain multichannel transcoder of the P frame sequence of AVS multichannel code stream by P frame sequence multichannel output module;
The macro block that adopts intraframe coding in the B frame in the multichannel intermediate code flow data is carried out compression domain multichannel transcoder in the B frame frame that parallelly compressed domain code conversion handles;
The macro block that adopts interframe encode in the B frame in the multichannel intermediate code flow data is carried out parallelly compressed domain code conversion handle, and export the B frame interframe compression domain multichannel transcoder of the B frame sequence of AVS multichannel code stream by B frame sequence multichannel output module;
Multichannel bit stream data after the conversion of the above-mentioned multichannel transcoder whole fast transcoding that walk abreast is handled, at last the whole quick multichannel transcoder by multi-channel A VS output module output multi-channel A VS code stream.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200820146470 CN201282535Y (en) | 2008-08-18 | 2008-08-18 | Device for converting H.264 to AVS video code stream |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200820146470 CN201282535Y (en) | 2008-08-18 | 2008-08-18 | Device for converting H.264 to AVS video code stream |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201282535Y true CN201282535Y (en) | 2009-07-29 |
Family
ID=40929400
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200820146470 Expired - Lifetime CN201282535Y (en) | 2008-08-18 | 2008-08-18 | Device for converting H.264 to AVS video code stream |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201282535Y (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102098502A (en) * | 2009-12-14 | 2011-06-15 | 无锡中星微电子有限公司 | Method and device for converting coding formats |
| CN102186082A (en) * | 2011-05-13 | 2011-09-14 | 哈尔滨工业大学 | H.264 protocol based optimized decoding method for intra-frame coding compression technology |
| CN101909211B (en) * | 2010-01-04 | 2012-05-23 | 西安电子科技大学 | H.264/AVC high-efficiency transcoder based on fast mode judgment |
-
2008
- 2008-08-18 CN CN 200820146470 patent/CN201282535Y/en not_active Expired - Lifetime
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102098502A (en) * | 2009-12-14 | 2011-06-15 | 无锡中星微电子有限公司 | Method and device for converting coding formats |
| CN102098502B (en) * | 2009-12-14 | 2015-06-03 | 无锡中星微电子有限公司 | Method and device for converting coding formats |
| CN101909211B (en) * | 2010-01-04 | 2012-05-23 | 西安电子科技大学 | H.264/AVC high-efficiency transcoder based on fast mode judgment |
| CN102186082A (en) * | 2011-05-13 | 2011-09-14 | 哈尔滨工业大学 | H.264 protocol based optimized decoding method for intra-frame coding compression technology |
| CN102186082B (en) * | 2011-05-13 | 2014-04-16 | 哈尔滨工业大学 | H.264 protocol based optimized decoding method for intra-frame coding compression technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101621687B (en) | Methodfor converting video code stream from H. 264 to AVS and device thereof | |
| TWI626842B (en) | Motion picture coding device and its operation method | |
| CN101946515B (en) | Two pass quantization for cabac coders | |
| US8582904B2 (en) | Method of second order prediction and video encoder and decoder using the same | |
| JP6120707B2 (en) | Video encoding apparatus and operation method thereof | |
| US20090103622A1 (en) | Method and system for determining a macroblock partition for data transcoding | |
| CN101247525B (en) | Method for improving image intraframe coding velocity | |
| CN105723707A (en) | Color residual prediction for video coding | |
| CN103891279A (en) | Most probable transform for intra prediction coding | |
| KR20050112130A (en) | Video transcoding | |
| CN1658673A (en) | Video compression coding-decoding method | |
| CN101568033A (en) | In-loop adaptive wiener filter for video coding and decoding | |
| CN108028934A (en) | Improved video flowing switching and random-access method and system | |
| CN102447907A (en) | Video sequence coding method aiming at HEVC (High Efficiency Video Coding) | |
| CN103442228B (en) | Code-transferring method and transcoder thereof in from standard H.264/AVC to the fast frame of HEVC standard | |
| CN104038764A (en) | H.264-to-H.265 video transcoding method and transcoder | |
| CN101984665A (en) | Video transmission quality evaluating method and system | |
| CN104322066B (en) | Block interleaved handling ordered for decoding video data | |
| CN102647593A (en) | AVS (Audio Video Standard) intra mode decision method and AVS intra mode decision device | |
| CN110351552B (en) | Fast coding method in video coding | |
| WO2021120614A1 (en) | Secondary coding optimization method | |
| CN113965763A (en) | Cross-codec coding optimization for video transcoding | |
| CN201282535Y (en) | Device for converting H.264 to AVS video code stream | |
| CN102137257B (en) | Embedded H.264 coding method based on TMS320DM642 chip | |
| CN102714717A (en) | Low-cost video encoder |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee | ||
| CP02 | Change in the address of a patent holder |
Address after: Nanshan District Nanshan Road Shenzhen city Guangdong province 518054 Nanshan Industrial Zone No. 1124 Nanyou fourth smart grid industrial park 807 Patentee after: Shenzhen Tieyue Electric Co., Ltd. Address before: 518067, F, 13, sea view Plaza, 18 Prince Edward Road, Shekou, Guangdong, Nanshan District, Shenzhen Patentee before: Shenzhen Tieyue Electric Co., Ltd. |
|
| DD01 | Delivery of document by public notice |
Addressee: Shenzhen Tieyue Electric Co., Ltd. Document name: Notification to Pay the Fees |
|
| DD01 | Delivery of document by public notice |
Addressee: Shenzhen Tieyue Electric Co., Ltd. Document name: Notification of Termination of Patent Right |
|
| C56 | Change in the name or address of the patentee | ||
| CP02 | Change in the address of a patent holder |
Address after: Dragon Street Longgang Huangge Road District of Shenzhen City, Guangdong province 518054 No. 441 Longgang Tianan cyber Park No. 1 building block B B804 Patentee after: Shenzhen Tieyue Electric Co., Ltd. Address before: Nanshan District Nanshan Road Shenzhen city Guangdong province 518054 Nanshan Industrial Zone No. 1124 Nanyou fourth smart grid industrial park 807 Patentee before: Shenzhen Tieyue Electric Co., Ltd. |
|
| CX01 | Expiry of patent term | ||
| CX01 | Expiry of patent term |
Granted publication date: 20090729 |