Method and apparatus for encoding and decoding audio and video data

Abstract

The invention discloses a method and a device for encoding and decoding audio and video data. The method comprises the following steps: the fixed-point operation data are input; the input data are calculated, wherein, the bit length of an operational unit is selected and the data operation accuracy is distributed through an overflow probability operation according to the encoding and decoding rule, the data area and the data distribution; whether the overflow occurs in an operation result is determined by a marker bit and an algorithm for determining the overflow; and the operation result is output under the condition that the overflow does not occur. By adopting the method, the data operational precision is improved and higher operational speed is realized, thereby the method is the significant improvement for the conventional encoding and decoding technology of audio and video.

Description

technical field [0001] The present invention relates to the field of audio and video encoding and decoding, in particular to a method and device for encoding and decoding audio and video data. Background technique [0002] In audio and video codec processing, the accuracy of fixed-point arithmetic data conversion is a very important issue. Insufficient precision directly affects the quality of audio or image. What's more, sometimes the lack of precision can add up, such as drift caused by IDCT precision mismatch. Therefore, in many cases, a certain calculation accuracy must be guaranteed. However, because the data needs to perform a series of calculations, there may be overflow problems, and high precision often requires the use of higher computing units, which usually brings about an increase in cost. Especially for some devices with lower bit-length arithmetic units, traditional methods often fail to balance precision and overflow, and use software to process data that ...

AI Technical Summary

Problems solved by technology

However, because the data needs to perform a series of calculations, there may be overflow problems, and high precision often requires the use of higher computing units, which usually leads to an increase in cost

Especially for some devices with lower bit-length arithmetic units, traditional methods often fail to balance precision and overflow, and use software to process data that exceeds the bit length of the arithmetic unit itself (such as using a 16-Bit CPU to process a large number of 32-Bit data, especially multiplication and division) method operation) will greatly reduce the processing speed of the operation

[0003] Also, in practice, a design with slightly less accuracy than theoretically required results in a very low chance of overflow

This is because, first of all, the distribution of a lot of input data within the legal value is uneven, basically concentrated in a small interval, and few of them reach or approach the allowable value. For example, in a certain application, the theory of IDCT operation The input value is [-1024, 1023], but the actual input value is basically concentrated within [-32, 32], and processing the data in the latter range can often reduce the accuracy of the data in the previous range.

Secondly, a lot of data has to be processed multiple times, which causes overflow to occur only when multiple data are in the extreme case, which further reduces the probability of overflow

Again, sometimes, certain ranges of data can be processed using simpler and faster methods, such as table lookup, etc., but the full range of data often cannot use this method

Images