数据驱动的AVS3像素域最小可觉差预测模型  

Just Noticeable Distortion Prediction Model of Data-Driven AVS3 Pixel Domain

作　　者：李兰兰 刘晓琳 吴珂欣 林丽群[1] 魏宏安[1] 赵铁松 LI Lanlan;LIU Xiaolin;WU Kexin;LIN Liqun;WEI Hongan;ZHAO Tiesong(Fujian Key Lab for Intelligent Processing and Wireless Transmission of Media Information,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,福建省媒体信息智能处理与无线传输重点实验室,福州350108

出　　处：《数据采集与处理》2021年第1期53-62,共10页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金面上(61671152)资助项目;国家自然科学基金青年科学基金(61901119)资助项目;福建省自然科学基金(2019J01222)资助项目。

摘　　要：AVS3作为中国第三代国家数字音视频编码技术标准,在消除视频时域/空域冗余信息方面发挥了重要的作用,但在消除感知冗余方面仍存在进一步优化的空间。本文提出一种数据驱动的AVS3像素域最小可觉差(Just noticeable distortion,JND)预测模型,在尽量保证视觉主观质量的前提下,对AVS3视频编码器进行优化。首先基于主流的大型JND主观数据库,获取符合人眼视觉特性的像素域JND阈值;然后基于深度神经网络构建像素域JND预测模型;最后通过预测的像素域JND阈值建立残差滤波器,消除AVS3的感知冗余,降低编码比特率。实验结果表明,与AVS3的标准测试模型HPM5.0相比,在人眼主观感知质量几乎无损的情况下,所提出的像素域JND模型最高可节省21.52%的码率,平均可节省5.11%的码率。The hybrid coding framework of the third generation audio and video coding standard(AVS3)plays an important role in eliminating redundant information in the video time domain/space domain,but needs to be further improved in eliminating perceptual redundancy and further improving coding performance.This paper proposes a just noticeable distortion(JND)prediction model of data-driven pixel domain to optimize AVS3 video encoder under the premise of ensuring the subjective quality of vision.Firstly,based on the current large subjective database of JND,the threshold of perceptive perception distortion in the pixel domain is obtained according to the human eye characteristics.Secondly,the pixel domain JND prediction model based on deep neural network is constructed.Finally,the residual filter established by the predicted pixel domain JND threshold is used to eliminate perceptual redundancy in AVS3 and reduce coding bitrate.The experimental results show that compared with the AVS3 standard test model HPM5.0,the proposed JND model can save up to 21.52%bitrate and an average of 5.11%bitrate.

关 键 词：视觉感知特性 最小可觉差预测模型 AVS3 残差滤波 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]