HEVC intra-frame coding compression performance optimization research combined with convolutional neural network

Abstract

The invention provides HEVC intra-frame coding compression performance optimization research combined with a convolutional neural network. The HEVC intra-frame coding compression performance optimization research integrates a down-sampling process, an HEVC coding and decoding process and an up-sampling and quality enhancement process. In order to efficiently extract the structural features of video frames, two convolutional neural networks are integrated in the proposed compression framework. According to the invention, a down-sampling convolutional neural network (DwSCNN, Down Sampling CNN) is provided to replace bicubic down-sampling; detail information is reserved while the resolution is effectively reduced; a more compact low-resolution video sequence is obtained; the low-resolution video sequence is subjected to further data volume compression through HEVC (High Efficiency Video Coding) intra-frame coding, and a quality-enhanced convolutional neural network (PPCNN,Post ProcessingCNN)) is provided to improve the quality-reduced video sequence which is recovered to the original resolution after decoding. The method can be widely applied to the fields of digital movie shooting transmission, literary and sports activity live broadcast, remote education training, target detection and the like.

Description

technical field [0001] The present invention relates to video coding technical issues in the field of image communication, in particular to a research on optimization of HEVC intra-frame coding compression performance combined with convolutional neural network. Background technique [0002] In recent years, the popularization of smart mobile terminals has made people have higher and higher requirements for video resolution. In this context, a new generation of high-efficiency video coding standard H.265 / HEVC (High Efficiency Video Coding) came into being. H.265 / HEVC includes the latest video coding technology, which has better performance than previous standards, and can save about 50% bit rate compared with H.264 / AVC under the same coding quality. Nowadays, the shooting of high-definition digital movies used in satellites, smart homes and other fields is compressed by intra-frame coding. The application of HEVC intra-frame coding is becoming more and more extensive. Due to...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the high-performance video compression coding standard HEVC compression efficiency and performance can be improved, further research to solve the problem of high-efficiency video compression, this paper proposes a HEVC intra-frame coding compression performance optimization research combined with convolutional neural network, in a reasonable time and complex Within the range of degrees, control the amount of video transmission data as much as possible and improve the quality of reconstructed video, and integrate the process of compressing and encoding data into one

Images