文献类型：期刊文献

中文题名：基于通用串预测算法的AVS2屏幕混合内容帧间编码优化

英文题名：Universal String Prediction-Based Inter Coding Algorithm Optimization in AVS2 Mixed Content Coding

作者：赵利平[1,3];林涛[2];郭靖[2];周开伦[2]

机构：[1]绍兴文理学院计算机科学与工程系,浙江绍兴312000;[2]同济大学超大规模集成电路研究所,上海200092;[3]嘉兴学院数理与信息工程学院,浙江嘉兴314000

年份：2019

卷号：42

期号：10

起止页码：2190

中文期刊名：计算机学报

外文期刊名：Chinese Journal of Computers

收录：CSTPCD、、EI(收录号：20195307948588)、北大核心2017、Scopus(收录号：2-s2.0-85077032180)、CSCD2019_2020、北大核心、CSCD

基金：This work is partially supported by the Natural Science Foundation of Zhejiang Province (No.LY19F020015), the National Natural Science Foundation of China (Nos.61871289 and 61601200), the Public Service Technology Application Research Project of Shaoxing City (No.2018C10015), and the Natural Science Foundation of Shanghai (Nos.18ZR1440600 and 19ZR1461100).

语种：中文

中文关键词：高效视频编码;音视频编解码技术标准;屏幕混合内容编码;帧间编码;快速算法

外文关键词：high efficiency video coding;audio video coding standard;screen mixed content coding;inter coding;fast algorithm

中文摘要：我国第二代数字音视频编解码技术标准的屏幕与混合内容扩展版(AVS2-SMCC)中,对于全色度屏幕混合内容的编解码其采用融合4∶4∶4格式的通用串预测编码算法和传统4∶2∶0格式的块预测、变换编码算法的编解码框架.在此框架中,根据屏幕混合内容的内在帧间特性和不同帧间子模式的特点,该框架对帧间编码算法进行了优化:首先提出了支持4∶4∶4格式的帧间编解码算法;其次,提出了内容自适应的不同帧间子模式色度失真权重因子算法和帧间子模式预编码与编码树单元划分的提前终止算法;最后,提出了一种新的帧间第一受限串模式,即将通用串预测算法的第一受限串的参考位置扩展至参考图像队列中的第一帧.实验结果表明,对于AVS2-SMCC标准测试序列中的移动文本和图形类别序列,提出的算法和AVS2-SMCC参考软件RMD1.0相比,在有损低延迟配置(一种含帧间编码的配置)下,Y、U、V分量的平均BD-rate下降率分别可达14.35%、90.73%、88.85%,编码时间降低50.86%,解码时间增加15.24%.

外文摘要：With the rapid development of technologies in networking and thin-client devices,such as remote desktop,video conferencing with documents or slides sharing,television programs and many curriculum videos playing in the Internet that contain mixtures of camera-captured content(CC)and screen content(SC),efficient screen content coding(SCC)becomes a hot topic in multimedia applications and has attracted increasing researcher attention from both academia and industry.Two international video coding standards include efficient SCC capability.One is High Efficiency Video Coding(HEVC),the other is the second-generation of the Audio Video Coding Standard(AVS2).In recent years,AVS is developing an AVS2 Screen and Mixed Content Coding(SMCC)extension(AVS2-SMCC).In our previous work,a 4∶4∶4 format universal string prediction(USP)algorithm integrated with the tradition 4∶2∶0 format block prediction and transform coding framework is used to code 4∶4∶4 format mixed content screen.To fully exploit both local and non-local,both general and special,and both complex and simple matching patterns with a variety of sizes and/or shapes and/or positions in a wide range of commonly seen screen content,a USP approach and its key technologies with three modes:general string(GS)mode,constrained string 1(CS1)mode,and constrained string 2(CS2)mode are proposed.The three constrained string modes are implemented with one of the three types of strings:offset string,coordinate string,and unpredictable pixel,or its combination of them.When using USP algorithm to code a coding unit,one of the three constrained modes which gets the minimum rate distortion value is selected to code the CU.Experimental results show that,for text and graphics with motion of AVS2-SMCC test sequences,USP algorithm achieves significantly improved coding efficiency in the All Intra configuration when compared with the latest HEVC-SCC extension.In order to further improve the coding efficiency in the Low Delay configuration and reduce the complexity,in this framework,according to the inherent inter-frame characteristics of the mixed content and the different characteristics for different inter-sub-coding modes,an optimized inter coding algorithm which consists of a 4∶4∶4 format inter-coding algorithm,a content adaptive weighted chroma distortion algorithm for different inter-sub-coding modes and an early terminate algorithm for inter-sub-coding modes pre-coding and coding tree unit partition,and a newly inter constrained string 1 mode(the search area of CS1 mode extends to the first frame in the reference picture queue)are proposed.Compared with the reference software RMD1.0 of AVS2-SMCC,for the category of text and graphics with motion in the AVS2-SMCC test sequences,the proposed algorithm achieves the average BD-rate reduction up to 14.35%,90.73%and 88.85%respectively,in lossy low delay configuration,with a nearly 50.86%decrease in encoding runtime and 15.24%increase in decoding runtime.Especially for the video category of text and graphics with motion,the proposed algorithm achieves significantly improved coding efficiency not only in the All Intra configuration but also in the Low Delay configuration when compared with the HEVC-SCC extension.