H.264is the latest new generation of very low bit rate video coding standard adopted by ITU. The purpose of this article is to elaborateH.264, and its application in video conference is introduced.
keywords:H.264Video editingcodeMultiple framesVideo Conferencing
ITU-TandISO/IEC JTC1is currently the official organization that formulates video coding standards in the world.,ITU-Tstandards are called recommendations and namedisH.26xseriesH.261,H.263, etc.ISO/IECisMPEG-x,MPEG-1,MPEG-2,MPEG-4, etc.H.26xseries of standards are mainly used for real-time video communication, such as video conference, videophone, etc;MPEGseries of standards are mainly used for video storage.storage(DVD), video broadcasting, and video streaming (e. g.onInternet,DSLvideo, wireless video, etc.). In addition to joint openingH.262/MPEG-2standards, in most cases, the two organizations independently formulate relevant standards.1997,ITU-T VCEGandISO/IEC MPEGonce again cooperated and establishedJoint Video Team (JVT), committed to the development of a new generation of video coding standardH.264.19981, draft solicitation began;19999, the first draft was completed.;20015, its test model was developed.TML-8;20026month,JVTNo.5meetingH.264FCDplate;200212month,ITU-Twas formally adopted at the meeting in Japan.H.264standards, andon20035. The International Telecommunication Union named the systemisH.264/AVC, the International Organization for Standardization and the International Electrotechnical Commission saidis14496-10/MPEG-4 AVC.
H.264the same as the previous standard, alsoisDPCMA hybrid encoding mode of plus transform encoding. But it takes"regression base"simple design, without many options, getratioH.263 ++much better compression performance; The ability to adapt to various channels has been strengthened"network friends"are conducive to the processing of error codes and packet loss. The application target range is wide to meet the needs of different rates, different resolutions and different transmission (storage) occasions.
technology, it concentrates the advantages of previous standards and absorbs the experience accumulated in standard-setting.andH.263 v2(H.263 +)orMPEG-4simpleclass(Simple Profile),H.264When using the best encoder similar to the above coding method, it can save at most at most code rates.province50%bit rate.H.264can continuously provide high video quality at all code rates.H.264can work in low latency mode to adapt to real-time communication(such as a video conference), at the same time, it can work well in applications without delay limit, such as video storage and server-based video streaming applications..H.264provides the tools needed to deal with packet loss in the packet transmission network and the tools to deal with bit errors in the error prone wireless network.
at the system level,H.264proposed a new concept in video codinglayer(Video Coding Layer, VCL)and network extractionlayer(Network Abstraction Layer, NAL). The former is the expression of the core compressed content of video content, and the latter is the expression of delivery through a specific type of network. This structure facilitates the encapsulation of information and better priority control of information.H.264is as follows1.1 H.264system block diagram
Intra coding is used to reduce the spatial redundancy of images. In order to mentionH.264the efficiency of intra-frame coding, the spatial correlation of adjacent macroblocks is fully utilized in a given frame, and adjacent macroblocks usually contain similar attributes. Therefore, when encoding a given macro block, it is first possible to predict based on the surrounding macro block (typically based on the macro block in the upper left corner, because this macro block has been encoded), and then to The difference between the predicted value and the actual value is encoded, so that the code rate can be greatly reduced compared to directly encoding the frame.
H.264mentionsupply6modeline4 ×4Pixel macroblock prediction1DC Predictionand5direction prediction, such2. In the figure, adjacent blocksAI9pixels have been encoded and can be used to predict. If we choose modulo4, then,a,B,c,d4pixels are predicted to beandEequal values,e,f,g,h4pixels are predicted to beandFequal values, for flat areas with little spatial information in the image,H.264also support16 × 16.2Intra-frame encoding mode
2.Inter-frame prediction encoding
Inter-frame prediction coding uses temporal redundancy in continuous frames for motion estimation and compensation.H.264's motion compensation supports most of the key features in previous video coding standards, and flexibly adds more functions, except for supportingPframe,BOut-of-frame,H.264also supports a new type of inter-stream transmissionframe--SPframes, such3. Package in code streamSPframe, it can quickly switch between streams with similar content but different bit rates. At the same time, it supports random access and fast playback modes.3 SP-Frame descriptionH.264is estimated to be4features.
(1))Macroblock segmentation of different sizes and shapes
for each16 × 16pixel macroblocks can be of different sizes and shapes.,H.264branch7mode, such4. The motion compensation of the small block mode improves the performance for the processing of motion details, reduces the block effect, and improves the quality of the image.4Macroblock segmentation method
(2))High Precision Sub-pixel Motion Compensation
inH.263uses half-pixel precision motion estimation, whileinH.264can be used1/4or1/8motion estimation of pixel accuracy. Under the condition that the same accuracy is required,H.264make1/4or1/8the residual error after motion estimation of pixel accuracyratioH.263the residual error after using half-pixel precision motion estimation is small. In this way, at the same accuracy,H.264The bit rate required in inter-frame encoding is smaller.
H.264provides an optional multi-frame prediction function. During inter-frame encoding, you can5different reference frames provide better error correction performance, which can improve the video image quality. This feature is mainly used in the following situations: periodic motion, translational motion, and changing the camera lens back and forth between two different scenes.
H.264defines a filter that adaptively removes the block effect, which can handle the horizontal and vertical block edges in the prediction loop, greatly reducing the block effect.
in transformation,H.264uses a baseon4 ×4the similarity of pixel blocksonDCTtransformation, but uses a spatial transformation based on integers. There is no inverse transformation. There is an error problem because of the trade-off. The transformation matrix is such5. Compared with floating-point operations, the wholeDCTtransformation will cause some additional errors, but becauseisDCTquantization after transformation also has quantization errors. In contrast, the wholeDCTtransformation has little influence. In addition, the wholeDCTtransformation also reduces the amount of computation and complexity, which is conducive to directional determination.pointDSPtransplantation.
H.26432different quantization steps.andH.263have31quantization steps are very similar,inH.264, the step size is12.5%is progressive, not a fixed constant.
inH.264, there are also two ways to read the transformation coefficients: the wordshape(Zigzag)Scan and double scan, such6. In most cases, simple zigzag scanning is used. Double scanning is only used in blocks with smaller quantization levels, which helps to improve coding efficiency.6How to Read Transformation Coefficient
The last step in video coding processing is entropy coding,inH.264: general variable length coding(UVLC) and text-based adaptive binary arithmetic coding(CABAC).
inH.263and other standards, different data types, such as transform coefficients and motion vectors, are usedVLCcode table.H.264UVLCcode table provides a simple method. No matter what type of data the symbol expresses, it uses a unified variable word long code table. Its advantage is simplicity; the disadvantage is that a single code table is derived from the probability statistical distribution model, without considering the correlation between coding symbols, and the effect is not very good at medium and high bit rates.
Therefore,H.264CABACMethod. Arithmetic coding enables both encoding and decoding to use all syntactic elementselement(transformation coefficient, motion vectorquantity)probability model. In order to improve the efficiency of arithmetic coding, the basic probability model can adapt to the statistical characteristics that change with video frames through the process of content modeling. Content modeling provides a conditional probability estimate of the encoded symbol, using the appropriate content model, the correlation between symbols can be removed by selecting the corresponding probability model of the encoded symbol that is currently adjacent to the encoded symbol, and different syntactic elements usually maintain different models.
IV,H.264in Video Conference
At present, most video conference systems adoptH.261orH.263Video encoding standard,andH.264, such that at the same rate,H.264canratioH.263minus50%. In other words, even if the user only benefits384kbit/sbandwidth, you can enjoyacceptH.263lower heightda768kbit/s's High Quality Video Service.H.264not only helps to save huge expenses, but also improves the efficiency of resource use, and at the same time enables video conferencing services of commercial quality to have more potential customers.
At present, a few manufacturers have announced their video conferencing products.H.264agreement, manufacturers are committed to the generalandH.264this new industry standard. As other video conferencing solution vendors follow their lead, we are bound to be able to fully implementtestH.264Advantages of video service.