基于深度学习的视频压缩综述  

An Overview of Video Compression Based on Deep Learning

作　　者：严昊 YAN Hao(College of Physics and Information Engineering,Fuzhou University,Fuzhou 350108,China)

机构地区：[1]福州大学物理与信息工程学院,福建福州350108

出　　处：《电视技术》2024年第1期216-219,共4页Video Engineering

摘　　要：近年来,深度神经网络以其强大的非线性建模能力,促进了视频压缩领域的发展,促使基于深度学习的视频压缩成为一个备受关注的研究领域。介绍基于深度学习的端到端视频压缩技术及其发展,重点介绍和总结现有的端到端视频压缩框架,并分析未来的端到端视频压缩发展趋势。In recent years,deep neural network,with its powerful non-linear modeling ability,has promoted the development of video compression,which makes video compression based on deep learnnig become a research field.This paper introduces the end-to-end video compression technology based on deep learning and its development,especially introduces and summarizes the existing end-toend video compression framework,and analyzes the future trend of end-to-end video compression.

关 键 词：深度学习 视频压缩 神经网络 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP18[自动化与计算机技术—计算机科学与技术]