首页 > 论文 > 激光与光电子学进展 > 55卷 > 7期(pp:71101--1)

基于三维卷积神经网络的无参考视频质量评价

No-Reference Video Quality Assessment Based on Three-Dimensional Convolutional Neural Networks

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为了在不借助参考视频的条件下准确评价失真视频质量,提出一种应用三维卷积神经网络提取失真视频时空域特征的通用型无参考视频质量评价算法。在视频质量库上训练卷积神经网络模型3D ConvNets,使3D ConvNets学习到与视频失真程度相关的特征;应用3D ConvNets对输入的失真视频进行特征提取,对提取得到的质量特征先后进行L2范数规则化和主成分分析以防止过拟合并去除冗余特征;使用线性支持向量回归根据视频质量特征预测失真视频的质量分数。实验结果表明,本文算法能够较为准确地评价多种视频失真类型,并且在更换测试视频库后依然保持较高的评价准确度,同时算法评价视频质量的计算复杂度极低。

Abstract

In order to assess the quality of distorted videos accurately without reference videos, a universal no-reference video quality assessment algorithm is proposed, which applies three-dimensional (3D) convolutional neural networks to extracting spatiotemporal features of distorted videos. Firstly, the convolutional neural network model 3D ConvNets is trained on the video quality database, and then the features related to video distortion degree are learned. Then, 3D ConvNets is used to extract features of the input distorted video, after which L2-normalization and principal component analysis are performed to prevent overfitting and eliminate redundancy. Finally, linear support vector regression is used to predict quality score of the distorted video based on video quality features. The experimental results show that the proposed algorithm can assess video quality accurately across different kinds of distortion, and it still maintains a high level of accuracy when the test video database is changed. Last but not least, the computational complexity of quality assessment process is extremely low for the proposed algorithm.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TN919.8

DOI:10.3788/lop55.071101

所属栏目:成像系统

收稿日期:2017-11-27

修改稿日期:2017-12-17

网络出版日期:--

作者单位    点击查看

张淑芳:天津大学电气自动化与信息工程学院, 天津 300072
郭志鹏:天津大学电气自动化与信息工程学院, 天津 300072

联系人作者:联系作者(shufangzhang@tju.edu.cn)

备注:张淑芳(1979—),女,博士,副教授,硕士生导师,主要从事图像处理方面的研究。E-mail: shufangzhang@tju.edu.cn

【1】Hou C P, Ma T T, Yue G H, et al. Multiply-distorted image quality assessment based on high-order phase congruency[J]. Laser & Optoelectronics Progress, 2017, 54(7): 071001.
侯春萍, 马彤彤, 岳广辉, 等. 基于高阶相位一致性的混合失真图像质量评价[J]. 激光与光电子学进展, 2017, 54(7): 071001.

【2】Zhang Y, Jin W Q. Assessment method of fusion image quality in wavelet domain structural similarity[J]. Chinese Journal of Lasers, 2012, 39(s1): s109007.
张勇, 金伟其. 小波域结构相似度融合图像质量评价方法[J]. 中国激光, 2012, 39(s1): s109007.

【3】Saad M A, Bovik A C, Charrier C. Blind prediction of natural video quality[J]. IEEE Transactions on Image Processing, 2014, 23(3): 1352-1365.

【4】Li X, Guo Q, Lu X. Spatiotemporal statistics for video quality assessment[J]. IEEE Transactions on Image Processing, 2016, 25(7): 3329-3342.

【5】Negi P S, Labate D. 3-D discrete shearlet transform and video processing[J]. IEEE Transactions on Image Processing, 2012, 21(6): 2944-2954.

【6】Li Y, Po L M, Cheung C H, et al. No-reference video quality assessment with 3D shearlet transform and convolutional neural networks[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2016, 26(6): 1044-1057.

【7】Wang C F, Su L, Zhang W G, et al. No reference video quality assessment based on 3D convolutional neural network[J]. Journal of Software, 2016, 27(2): 103-112.
王春峰, 苏荔, 张维刚, 等. 基于3D卷积神经网络的无参考视频质量评价[J]. 软件学报, 2017, 27(2): 103-112.

【8】Chen J, Jiang H, Liu T D, et al. Optimization for Raman fiber amplifiers based on least squares support vector regression model[J]. Acta Optica Sinica, 2015, 35(11): 1123004.
陈静, 江灏, 刘暾东, 等. 基于最小二乘支持向量回归模型的拉曼光纤放大器优化设计[J]. 光学学报, 2015, 35(11): 1123004.

【9】Tran D, Bourdev L, Fergus R, et al. Learning spatiotemporal features with 3D convolutional networks[C]∥Proceedings of the IEEE International Conference on Computer Vision, 2015: 4489-4497.

【10】Seshadrinathan K, Soundararajan R, Bovik A C, et al. Study of subjective and objective quality assessment of video[J]. IEEE Transactions on Image Processing, 2010, 19(6): 1427-1441.

【11】Vu P V, Chandler D M. ViS3: an algorithm for video quality assessment via analysis of spatial and spatiotemporal slices[J]. Journal of Electronic Imaging, 2014, 23(1): 013016.

【12】Wang Z, Bovik A C, Sheikh H R, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612.

【13】Seshadrinathan K, Bovik A C. Motion tuned spatio-temporal quality assessment of natural videos[J]. IEEE Transactions on Image Processing, 2010, 19(2): 335-350.

【14】Pinson M H, Wolf S. A new standardized method for objectively measuring video quality[J]. IEEE Transactions on Broadcasting, 2004, 50(3): 312-322.

引用该论文

Zhang Shufang,Guo Zhipeng. No-Reference Video Quality Assessment Based on Three-Dimensional Convolutional Neural Networks[J]. Laser & Optoelectronics Progress, 2018, 55(7): 071101

张淑芳,郭志鹏. 基于三维卷积神经网络的无参考视频质量评价[J]. 激光与光电子学进展, 2018, 55(7): 071101

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF