"Transfer from: http://www.oschina.net/news/47285/H265
When the author downloads HD movies, if you see H.264 in the encoding format, you will feel more assured because the picture quality in this coding format is good. However, such movie files are very large, in my country's current network status, downloading a favorite HD movie really requires unparalleled patience. Perhaps someone thinks to go to the video website to see HD, but in the eyes of picky shadow, online video is far from high-definition standard. However, this situation may have changed because the H.265 standard of higher quality video is to be transmitted under limited bandwidth, and the video coding is now added. H.265 is also known as Hevc (High Efficieny Video Coding), is a new generation of video compression standards developed by international standardized organizations and ITU organizations, mainly for the application of HD digital TV and video codec system. In August last year, Ericsson Company launched the first H.265 encoded decoder. After 6 months, ITU (ITU) officially approved the standard. The goal of H.265 is that the encoding efficiency is 50% higher than H.264, that is, under the same image quality conditions, the target code rate drops to 50% of H.264, and the bandwidth is only half. This means that our mobile devices will be able to play all HD (1080P) videos directly online. The H.265 standard also supports video applications from SQICF (128 × 96) to 8K Ultra HD (8192 × 4320). Before you understand the advantages of H.265, we may wish to review H.264 first. About H.264 H.264 is a Height Compressed Digital Video, a combined video group (JVT, Joint Video Team), composed of ITU-T video coding expert group (VCEG) and ISO / IEC Dynamic Image Expert Group (MPEG) Decoder standard. H.264 is one of the blue-ray codec standards, and all Blu-ray players must be able to decode H.264. It is also widely used in various high-precision video recording, compression, and release. More importantly, Apple has finally abandoned Adobe's VP6 encoding, chosen H.264, allowing this standard with countless iPad and iPhone into thousands of households, eventually became the absolute hegemony in the high-definition video coding field. H.264 can be logically divided into two layers: video coding layer (VCL: Video Coding Layer) is responsible for efficient video content, network abstraction layer (NAL: NETWORK ABSTRACTION LAYER) is responsible for data on the appropriate way of network Package and transfer. The VCL layer data is an output of the encoding process, which represents a compressed encoded video data sequence. H.264 uses DCT transformation encoding the mixed coding structure of D PCM differential encoding, but also increases, such as multi-mode motion estimation, intra prediction, multi-frame prediction, content-based beam coding, 4 × 4 two-dimensional integer transformation, etc. The encoding method improves coding efficiency. NAL is responsible for encapsulating or mapping data using the segmentation format of the lower layer network, including the signaling of the group frame, the logical channel, the utilization of timing information, or the sequence end signal, and the like. NAL supports video in the transport format on the circuit switched channel, supports video on the Internet using RTP / UDP / IP transmission format. H.264 includes tools for error-elimination, easy to compress video in easily errors, packet loss, enhance fault tolerance performance. Summarize, H.264 has the following characteristics of the compression standard: higher coding efficiency: Under the same image quality, the compression ratio of H.264 is more than 2 times the MPEG-2, which is MPEG- 4. 1.5 ~ 2 times. A 120-minute full HD movie can be compressed to about 20 g B, while the picture quality is superior to the MPEG-2 format. Relative to spread and store. Better network adaptability: H.264 can work in real-time communication applications (such as video conferencing) low delay mode, you can also work in a video storage or video stream server without delay, and solve the network well Transferring packet loss issues. Of course, the performance improvement of H.264 is obtained at the expense of increasing the complexity of decoding calculations, so some "older machine" plays MPEG-2 HD video may have no pressure, but the video of playing H.264 will be quite tasty . But now, it is a tablet that has supported H.264 hard solution. The limitations of H.264 Every HD enthusiast has a good past, but H.264 can't stay with you forever. Because the current digital video is moving towards the following "metamorphosis" direction: 1, higher clarity: 720p is outdated, 1080p also showed the videos of the old, 4K, 8K, is coming to us; 2, higher Frame rate: The cinema does not satisfy the frame rate of 24fps, Peter Jackson shoots "Hobbit" uses a 48FPS frame rate. Digital video is even upgraded towards 60fps, 120fps or even 240fps; 3. Higher compression ratio: clear screen and limited bandwidth and storage space will always be contradictory. Therefore, obtaining the best video effect under limited bandwidth and storage space, is a relentless pursuit of all relevant personnel in the high-definition field. In the face of these three "metamorphosis" conditions, if we have been holding H.264, there will be some problems: the explosive growth of the number will result in prediction mode, sports vector, reference frame for encoding macroblocks. The codewords occupied by macroblock-level parameter information such as index and quantization level, resulting in a decrease in codewords for encoding residual part. 2. Due to the increase in resolution, the information of the image content represented by a single macroblock is greatly reduced, causing an adjacent 4 × 4 or 8 × 8 block to perform integer transformation, the low frequency coefficient is greatly improved, and there is a lot of redundant Remain. 3. Due to the increase in resolution, the amplitude of the motion vector of the same movement will increase significantly, and the compression ratio of the method for predicting the motion vector in H.264 will gradually decrease. 4. Some key algorithms of H.264 are, for example, two context-based entropy codes based on CAVLC and Cabac, and require serial encoding, and less performance is relatively low. The current GPU, DSP, FPGA, and ASICs and other parallelization are very high. This serial feature of H.264 is increasinglycoming a bottleneck restricting computing performance. H.265 is to solve these problems. The H.265 parse the encoding architecture of the H.265 / HEVC is similar to the architecture of H.264 / AVC, and also includes an intra prediction, inter prediction, conversion, quantization, demineral block filter, entropy encoding and other modules. In order to improve the compression coding efficiency of HD video, H.265 proposes a large-size four-tree coding architecture and adopts a Coding Unit, Cu, predictive unit (PRANSFORM UNIT, TU). Three basic units perform the entire encoding process. Under this mixed encoding frame, H.265 has a large number of technological innovations, such as the split structure and residual coding structure based on large-size four-tree blocks, multi-angle intra prediction technology, motion estimation fusion technology, high-precision movement Compensation technology, adaptive loop filtering technology, and semantic entropy encoding techniques. Cu is similar to macroblock in H.264 / AVC. Each macroblock size in H.264 is fixed 16 × 16 pixels, while the H.265 CU can choose from minimum 8 × 8 to max 64 × 64 pixels. Take FIG. 4 as an example, the detailed area (such as the red portion of the vehicle body and the gray portion of the ground) divided by the CU, less data, less data; more than a small amount of CU is small Moreover, the encoded data is more, so that the image has a focus on the image, and the coding efficiency is improved. The PU is a basic unit for prediction. The H.265 uses the PU to implement the prediction process of each Cu unit. The PU size is limited to its Cu, which may be, for example, a block of 64 × 64 pixels, or a rectangle of, for example, 64 × 32 pixels. There is also a new asymmetric motion segmentation, AMP scheme, that is, to divide the coding unit into a prediction block that is inconsistent with two dimensions. This prediction method takes into account the possible texture distribution, which can effectively improve the prediction efficiency of large size blocks. T u is a basic unit for conversion and quantization. H.265 breaks the original transform dimensional limit, supports 4 × 4 to 32 × 32 encoding transformations, and transforms and quantizes the basic unit with TU. In order to improve the encoding efficiency of large-size coding units, DCT transform also uses a four-fork type transform structure. The separation of the three units of CP, PU, TU makes the transformation, predicts, and encoding the various processing links more flexible, and also facilitates the division of each link to more conform to the texture characteristics of the video image, which facilitates the completion of each unit. . The basic framework of H.265, the basic framework of intra prediction is basically the same as H.264: the correction of the neighbor block reconstruction pixels perform intra prediction from the neighbor block, select the predicted motion vector, support more Reference frame prediction, etc. However, the intra prediction mode of H.265 supports 33 directions (H.264 only supports only 8), and provides better motion compensation processing and vector prediction methods. At the same time, in order to adapt to the current parallel level of chip architecture, H.265 introduces many parallel operations to improve coding decoding efficiency. Repeated quality comparison test has shown that under the same image quality, the video encoded by H.265 will be reduced by about 40% by the video volume encoded by H.264. At the same time, in the case of less than half the rate of size, the quality of the H.265 encoded video can also be approximately even better than the H.264 encoded video. Although this conclusion is subjective, it is also very inspiring. The current H. 265 standard has three modes: Main, Main 10 and Main STILL PICTURE. The main model supports 8Bit color depth (ie, 256 chromasings each having a total of 167 million colors), and the main 10 mode supports 10 bit color depth, which will be used in UHDTVs (UHDTV). H.265 has a high level of quality on HD development, which is a high quality of H.265. We can now estimate that it has the following effects on the development of future digital videos: Let HD video surveillance become a reality: HD video is popular in security monitoring, but also brings a sharp rise in storage costs and space. The high compression ratio of H.265 can save half of the storage costs of H. 265, significantly reduced the storage cost of the video. Let 4K ultra HD TV faster: At present, 4K super HD TV products have been a lot, and some products have been as low as 45,000. There is no too much pressure in many families, but 4K sources are difficult to find, 4K TV is completely high-tech guns. With the maturity of H.265 related hardware and software, when a Blu-ray disc can accommodate a 4K film, it is the time when these 4K ultra HD TV is really playing. Let HD TV programs are more popular: current cable TV systems, the earlier MPEG-2 standard, even H.264 is not reached. The reason is the overall limitations of the transport network bandwidth, and the H.264 encoding also needs to pay the license fee. The H.265 is more conducive to transmission while ensuring the image quality, the same content, H.265 can reduce bandwidth consumption of about 70% to 80%. This is also possible to convince those stubborn radio and television companies to abandon MPEG-2, and use H.265 to easily propagate all HD 1080P TV programs under existing bandwidth conditions. For the purpose of saving bandwidth, satellite TV companies may be more active. Let go of HD video anytime, anywhere, possible: With the development of mobile devices, people need to access the required content and services anytime, anywhere. LTE and other 4G technology significantly improve the commercial feasibility of video services. The PPI and resolution of the mobile terminal display are getting higher and higher, plus H.265 high compression rate features, will further promote the popularity of high-definition or even super high-definition video under 4G networks. In addition, the so-called "HD" of the video website is far from high-definition standard. Taking the "Transformers 3: Moonlight" in the hands of the author, this is the full HD version of the Blu-ray original copy after the Blu-ray original format is used in X.264 format, and the average rate is 17.5MB. / s, the file is 18.9GB. This movie wants to watch directly online, and the network bandwidth must be greater than its average code rate. This is almost impossible to task. The author found on the Thunder to see the "Transformers 3" HD version and played directly through the webpage. Since the film's media information cannot be directly obtained, the author is estimated. Website requires that the bandwidth of HD video must reach 4MB. The 4MB bandwidth means that the upper speed is up to 500kb / s. The online video playback process is played while the download is played, and if the picture is smooth, the video file volume must be less than the current bandwidth download limit. The "Transformers 3" is 154 minutes, and the full-scale can be downloaded 4.7GB of files at 4MB bandwidth. Correspondingly, the average code rate is naturally less than 4Mb / s. 18.9GB of the full HD version is far away. Therefore, this is not a real HD video. At present, Thunder looks at the client's experience version of the H.265, and there is also "Transformers 3". By viewing media information, the video file is 2.55GB, the average rate of 2362kb / s. Subjective feels its picture quality exceeds the HD version of the web page, but there is still a big gap with 18.9GB HD version, after all, the code rate is more than 7 times. However, this gives at least one signal: if the code rate of the 4Mb / s is used to encode H.265, the picture quality gap will be greatly reduced. The problem behind the H.265 seems to be bright, but there are some problems with some problems to be solved. The equipment, including flat-screen TVs and HD players, can decode H.264 encoded files, while H.265 belongs to new technology, using H.265 encoded video file name HM10, its image quality and computational complexity and H .264Compared with a few times more, specially software and hardware requires corresponding support. Many players are currently not playing H.265 videos. At the same time, H.265 encoding also involves the issue of authorization, if commercialization must pay the corresponding authorization fees, then this fee is borne by the manufacturer or consumers? In addition to the price, what is the problem of consumers care about when they can buy related products. In terms of hardware, the author believes that the AMD, Ying Weida and other graphics manufacturers will stand up to stand up and support H.265. However, in the field of mobile equipment, some CPUs are powerful to mobile phones, MP4, etc., which can be soft decoded to H.265, may be quickly packaged into "support H.265 video decoding" to flicker consumers. But this support means high speed consumption of electricity (because the amount of operation is too large), and it is not true support, so please don't see "H.265" and rush. In the future, the H.265 standard will become a 4K and 8K resolution of ultra HDTV (UHDTV). But how should these video content be transmitted? What standards have been used? Sony and other media giants are ramping juice. The Blu-Ray Disc Association is also studying how to support the existing Blu-ray standard support 4K resolution video. Theoretically, H.265 can make Blu-ray discs to accommodate 4K movies, but it is necessary to make a new player to play, which is not a good thing for manufacturers and consumers. Despite this hindrance, we should still have a cautious attitude towards H.265. One thing is certain, H. 265 will significantly reduce bandwidth consumption in the same quality. We talked about the high-definition broadcast and TV and the popularity of 4K videos, which may no longer have some difficulties with H.265. X.264 is an open source project, based on H.264 encoding technology, free coding format for amateur market, is a subset of H.264, and only part of H.264 can be implemented. The X.264 is used for heavy compressed video content that flows on the network. "
Our other product:
