The H.264 is the same as the previous standard, or the mixed encoded framework, the AVS video standard uses a technical framework similar to H.264, including transformation, quantization, entropy coding, intra prediction, inter prediction, loop filtering The module. Their different core technologies include the following:
A transform and quantization
H.264 residual data using block-based transform coding, spatial redundancy is removed the original image, the ability to focus image on a small portion of the coefficients, the DC coefficient value is generally the largest, which can improve the compression ratio, enhanced anti interference. Previous standards generally use DCT transformation, this transformation is disadvantage mismatch phenomenon occurs, the original data after the transformation and inverse transformation will be a difference in recovery, since real number operation is also relatively large amount of computation. H.264-based integer transform is used in the 44.
AVS using 88 integer transform can be achieved without a mismatch in the 16-bit processors. High-resolution video images decorrelation conversion effective than 44, 64 using the quantization, can be adapted to different applications and service requirements of the streams and quality.
Second, intra prediction
H.264 and AVS technology are used for intra prediction mode, the prediction of the current block using neighboring pixels, using a variety of prediction modes representative of the spatial domain texture. H.264 luminance prediction blocks 44 and 16 × 16 block 2 kinds of prediction mode for a block of 4 × 4: +22.5 degrees from -135 degrees to apply a DC prediction direction of a total of nine kinds of prediction directions; for 16 × 16: there are four prediction directions. Chroma prediction is 8 × 8 block, there are 4 prediction modes, similar to the intra 16 × 16 prediction modes 4, wherein the DC mode is 0, a horizontal mode 1, mode 2 is a vertical, plane mode 3.
Third, the inter prediction
H.264 inter prediction coding using prediction mode based on the video frame and the motion compensated block, differs from the conventional standard block size of inter prediction that a wider range, the use of sub-pixel motion vectors and the use of multiple reference frames.
H.264 has a 16 × 16,16 × 8,8 × 16,8 × 8,8 × 4,4 × 8 and 4 × 4 macro block and a total of eight kinds of the sub-macroblock partition, whereas only 16 × 16,16 AVS × 8,8 × 16 8 × 8 and a total of four kinds of macro-block division manner.
H.264 supports a plurality of different reference frame for inter-predicted macroblocks and the sheet, AVS P frame in the reference frame may be used before and after each of a backward reference frames, B frames before using up two frames.
Fourth, entropy coding
H.264 entropy coding efficiency established based on the amount of information, one is unified variable length coding (UVLC) for all symbols to be encoded, and the other is based adaptive binary arithmetic coding (CABAC, Context-Adaptive Binary Arithmetic coding), greatly reducing the associated redundancy block coding, coding efficiency is improved. UVLC lower computational complexity, mainly for the encoding time is very critical applications, the disadvantage is low efficiency, high bit rate. CABAC is a highly efficient coding method, coding efficiency higher than 50% UVLC coding.
AVS entropy coding using adaptive variable length coding technique. In AVS entropy encoding process, all syntax elements and the residual data are mapped into a binary bit stream in the form of the exponential Golomb code.
Exponential Golomb code advantages: on the one hand, its relatively low hardware complexity, can be resolved in accordance with the formula codon closed, no look-up table; on the other hand, it may be determined flexibly according to a probability distribution of the elements in the coding order exponential Golomb K coding, if properly selected to be K, the encoding efficiency can approximate entropy.
Prediction residual transform block coefficients scanned form (level, run) of the string, level, RUN is not an independent event, and there is a strong correlation, level in the AVS, RUN jointly coded two-dimensional, and in accordance with the current level, different probability distributions run trend, adaptive to change the order of the exponential Golomb code.
Further, there is no SI in the AVS, SP frame. It can be said, AVS is developed on the basis of the H.264, H.264 absorbed the essence, but in order to bypass the troubled patent, but also had to give up some of the core algorithms for H.264. In the case in exchange for cost me a bit of coding efficiency is slightly reduced complexity was greatly reduced. , Reading the full text, the technology area
Tektron supports Amazon (AWS) media service, providing quality assurance for end-to-end video
IMEC is about to shock the first short-wave infrared (SWIR) band hyperspectral imaging camera
4K super high-definition home theater projector brings HD experience, full of fun
Video display system design based on unified calculation architecture technology
Apple TV 4K dismantling report: familiar modular components
Our other product:
