H.264 efficiency

H.264 takes video compression technology to the next level.

The H.264 standard introduces a new advanced intra prediction scheme to encode I-frames. This scheme can significantly reduce the size (in bits) of an I frame while maintaining high quality by successfully predicting small blocks of pixels within a macroblock within a frame. It does this by trying to find matching pixels between the previously encoded pixels that constrain the new 4×4 pixel block for internal encoding. By reusing already encoded pixel values, the bit size of the data can be significantly reduced. The new intra prediction is a key element of H.264 technology and has proven to be very effective. For comparison: even if only I-frames are used in H.264 streaming, the resulting file size will be much smaller than for Motion JPEG streaming.
Illustrations of how some intra-prediction modes can be used when encoding 4×4 pixels within one of the 16 blocks that make up a macroblock. Each of the 16 blocks within a macroblock can be encoded using different modes.

H.264 also improves block-based motion compensation for P and B frame encoding. The H.264 encoder can choose to search for matching blocks (up to sub-pixels

precision) of some or many sections within one or more reference frames. You can also adjust the size and shape of the blocks to improve your search for matches. In areas where matching blocks cannot be found within the frame of reference, internally coded macroblocks are used. The high degree of flexibility of block-by-block motion compensation in H.264 is effective in high-density video environments where image quality must be maintained for the application. Motion compensation is the most requested aspect of a video encoder, and the various ways and levels in which it can be implemented with an H.264 encoder help improve the efficiency of video compression.

With the H.264 standard, typical blocky objects seen in highly compressed Motion JPEG and MPEG (as opposed to H.264) videos can be reduced with the built-in unblocking filter. This filter automatically smooths the edges of the blocks to produce a nearly perfect unwrapped video image.

conclusion
H.264 represents a major step forward in video compression technology. This standard offers various technologies to achieve better compression efficiency through the use of more accurate intra prediction schemes, as well as more robustness. It opens up new possibilities to create advanced video encoders that can significantly improve image quality, increase frame rate and resolution while maintaining the same bit rate (compared to previous standards) or, conversely, provide video of the same quality at a lower bit rate.

H.264 is the first example of a collaboration between the International Telecommunication Union, the International Organization for Standardization and the International Electrotechnical Commission on International Model Video Compression Standards. Due to its flexibility, H.264 has found applications in areas as diverse as high definition DVD (such as Bluray), video streaming including HDTV streaming, online video storage (such as YouTube), 3G mobile phone. , in programs such as QuickTime, Flash, and the MacOS X operating system on Apple computers, as well as on video game consoles such as PlayStation 3. With the support of many industries and software development,

With the wider adoption of the H.264 format in network cameras, video encoders, and video management software, system designers and integrators must be confident that their chosen products and vendors are compliant with this new open standard. Today, network video equipment that supports H.264 and Motion JPEG is the ideal choice, offering the highest degree of versatility and integration.

H.264 Profiles and Levels

The joint group involved in defining the H.264 standard has focused on creating a simple and clear solution that minimizes options and parameters.

The most important aspect of this standard, as is the case with other video standards, is the provision of various capabilities within profiles (sets of algorithmic parameters) and levels (performance classes) to optimally support popular products and common formats. .

There are seven profiles in H.264, each of which is tailored to a specific application. Each profile determines which set of parameters the encoder can use and limits the complexity of the decoder implementation.

Network cameras and video encoders are likely to use a so-called baseline profile, designed primarily for use in areas with limited computing power. The baseline profile is best suited for use in modern real-time encoders embedded in network video equipment. This profile also provides low latency, which is an important requirement for video surveillance, and is essential for real-time control of pan, tilt and zoom of network PTZ cameras.

H.264 has 11 levels or degrees of limitation in terms of functionality, bandwidth, and memory requirements. Each level defines the bit rate and encoding rate in macroblocks per second for resolutions ranging from QCIF to HDTV and beyond. The higher the resolution, the higher the required level.