An ultra-low-latency ultra-high-definition video compression encoding and decoding method

Abstract

The invention belongs to the technical field of wireless multimedia communication, and discloses an ultra-high-definition video compression encoding and decoding method with ultra-low time delay. The encoding end performs input pre-processing on the input video image to remove the DC component; the pre-processed image is processed The sub-bands with different resolutions are obtained by asymmetric wavelet decomposition, and the transformed wavelet coefficients are proportionally extracted in the frequency domain. The processed wavelet coefficients are reasonably divided into regions, and the code rate control is performed on them to obtain the corresponding quantization parameters and entropy coding modes; entropy coding and code stream organization are performed on the quantized code group coefficients; at the decoding end, the obtained Perform header analysis on the code stream to obtain decoding parameters; perform entropy decoding to obtain wavelet coefficient values ​​and bit plane count positions, perform interpolation recovery after inverse quantization; perform inverse wavelet transform to obtain the original image data. While effectively reducing the complexity of compression and decompression algorithms, the present invention realizes high-efficiency compression and decompression and low-delay output of ultra-high-definition video.

Description

technical field [0001] The invention belongs to the technical field of wireless multimedia communication, and in particular relates to an ultra-low-latency ultra-high-definition video compression encoding and decoding method. Background technique [0002] At present, with the rapid development of video coding technology and display devices, video, as an efficient information transmission method, has been widely distributed in the Internet, TV broadcasting and various emerging media applications. The demand for ultra-low-latency ultra-high-definition video transmission in indoor scenarios is becoming more and more common, making indoor short-distance wireless high-speed multimedia transmission technology a research hotspot in recent years. The requirements of these applications are more stringent for indoor short-distance wireless high-speed multimedia transmission, requiring low delay, high speed, large capacity and high reliability transmission. [0003] The era of ultra-h...

AI Technical Summary

Problems solved by technology

Compared with the traditional HEVC intra prediction algorithm, it saves 4.4% of the code rate, but because of the use of fully connected neural networks and floating point operations in the multiplication operation, the complexity is extremely high

In fact, the improvement of the convolutional neural network coding algorithm for inter-frame prediction is because the designed neural network can reduce the block effect to improve the quality of inter-frame prediction, and at the same time, the complexity is extremely high, which is not conducive to low-latency compression coding.

[0005] Through the above analysis, the existing problems and defects of the existing technology are: the complexity of the existing high-definition / ultra-high-definition video compression coding algorithm is extremely high, and the delay of the coding and decoding loop is relatively large

[0006] The difficulty of solving the above problems and defects is as follows: the huge increase in the data volume of ultra-high-definition video brings great difficulties to the compression coding algorithm, and it is necessary to encode the data at the source end with a higher compression ratio

However, as the compression ratio increases, the image quality output by the source encoder will decrease, which cannot meet people's requirements for ultra-high-definition image quality.

At the same time, the source encoder adopts a variety of hybrid techniques to improve the compression ratio, which will greatly increase the complexity and encoding delay of the encoder, so it is necessary to compromise the compression ratio and quality in the actual algorithm design

[0007] The significance of solving the above problems and defects is: real-time multimedia applications in indoor scenes are very sensitive to encoding delays, and excessive encoding delays will affect the quality of user experience, causing discomfort to users during the experience process, and even in game competitions. time will bring great disaster

Images