Parallel decoding method and device

Abstract

Provided is a parallel decoding method and device. A macro block decoding operation comprises an entropy decoding operation and a post-decoding operation except the entropy decoding operation. The parallel decoding method comprises steps of: receiving a code stream from an encoder side, wherein the code stream is subjected to entropy coding by CABAC entropy coding technology; performing CABAC entropy decoding on macro blocks in the received code stream by using an individual CABAC entropy decoding threading; defining macro blocks capable of being subjected to the post-decoding operation at least partially on the basis of a CABAC entropy decoding result, a dependent relation between the intra-frame macro blocks and inter-frame macro blocks, and currently-available storage resources and computing resources; and performing the post-decoding operation on the defined macro blocks in parallel by using a parallel block decoding threading.

Description

technical field [0001] The present invention generally relates to . Background technique [0002] Traditional video coding standards, starting from H.261, after MPEG-1, MPEG-2, H.263, MPEG-4, H.264, etc., are all complicated encoders and relatively simple decoders; encoders can be freely Define the algorithm, and the decoder algorithm is fixed. The main reason for this is that the decoder is a mass consumer product, and the decoder is often implemented by a hardware chip. In order for the decoder to receive code streams compiled by different encoders, the code stream format must be standardized and consistent. Therefore, the decoder The decoding process is also standardized, and there is almost no room for modification or improvement, which actually increases the mass production quantity of the decoder and reduces the design and manufacturing cost of the decoder. [0003] However, as the codec standards become more and more complex, the decoders also become more and more c...

AI Technical Summary

Problems solved by technology

The disadvantage of this scheme is that 1) the decoding speed of the current closest frame has not been improved, and the first GOP is still very slow; The first GOP is still slow, and the decoding of many previous GOPs was wasted; 3) The cache requirements are very large, and the data of the following GOPs needs to be downloaded in advance when decoding online in real time, the network cache is enlarged, and more caches are needed at the same time Save the image data after GOP decoding is completed

The disadvantages of this scheme are: 1) It needs the cooperation of the encoder side, and the encoder must first divide the image into small images, otherwise the decoding end cannot be parallelized; 2) Since it is divided into small images, the small images cannot refer to each other, which seriously affects Coding efficiency, that is, the compression rate will become lower

The disadvantages of this scheme are: 1) It requires the cooperation of the encoding end. If the encoding end does not encode in parallel, the decoding end cannot be parallelized; there will be a little drop

The determination of this scheme is: since the execution time granularity of each block is already very small (millisecond level), and the time slice size scheduled by the operating system is generally about 20 milliseconds, so this kind of parallel design is very difficult and the hardware compatibility is poor.

Images