Quality of Experience Assessment for Free-viewpoint Video

IEEE international Symposium on Broadband Multimedia Systems and Broadcasting Pub Date : 2023-06-14 DOI:10.1109/BMSB58369.2023.10211612

Rongli Jia, Yuhong Zhang, Jun Xu, Wenjun Zhang, Li Song, Lin Li, Yanan Feng

{"title":"Quality of Experience Assessment for Free-viewpoint Video","authors":"Rongli Jia, Yuhong Zhang, Jun Xu, Wenjun Zhang, Li Song, Lin Li, Yanan Feng","doi":"10.1109/BMSB58369.2023.10211612","DOIUrl":null,"url":null,"abstract":"The evaluation of free-viewpoint video (FVV) quality is essential for improving the quality of experience (QoE). Prior deep video quality assessment (VQA) approaches for FVV typically focused on either spatial or temporal distortions and lacked a comprehensive assessment considering the two aspects. In this paper, we provide an end-to-end no-reference video quality assessment (NRVQA) model for FVV that predicts video quality scores based on both spatial and temporal features. It consists of a spatial feature perception module, a temporal motion feature perception module and a quality score fusion module. In order to provide a quality score that is highly relevant to the mean opinion score (MOS) from the subjective quality assessment experiment, the quality-related features in the spatial and temporal domains of FVV are effectively utilized and merged. Experimental results show that the PLCC and SRCC improved by 25.0% and 18.5%, respectively, compared to state-of-the-art method. Moreover, the ablation experiments demonstrate the importance of both spatial features and temporal motion features.","PeriodicalId":13080,"journal":{"name":"IEEE international Symposium on Broadband Multimedia Systems and Broadcasting","volume":"121 1","pages":"1-6"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE international Symposium on Broadband Multimedia Systems and Broadcasting","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BMSB58369.2023.10211612","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

The evaluation of free-viewpoint video (FVV) quality is essential for improving the quality of experience (QoE). Prior deep video quality assessment (VQA) approaches for FVV typically focused on either spatial or temporal distortions and lacked a comprehensive assessment considering the two aspects. In this paper, we provide an end-to-end no-reference video quality assessment (NRVQA) model for FVV that predicts video quality scores based on both spatial and temporal features. It consists of a spatial feature perception module, a temporal motion feature perception module and a quality score fusion module. In order to provide a quality score that is highly relevant to the mean opinion score (MOS) from the subjective quality assessment experiment, the quality-related features in the spatial and temporal domains of FVV are effectively utilized and merged. Experimental results show that the PLCC and SRCC improved by 25.0% and 18.5%, respectively, compared to state-of-the-art method. Moreover, the ablation experiments demonstrate the importance of both spatial features and temporal motion features.

查看原文本刊更多论文

免费视点视频的体验质量评估

自由视点视频(FVV)质量评价是提高视频体验质量的关键。以往的深度视频质量评估(VQA)方法通常侧重于空间或时间畸变，缺乏考虑这两个方面的综合评估。在本文中，我们为FVV提供了一个端到端的无参考视频质量评估(NRVQA)模型，该模型基于空间和时间特征预测视频质量分数。它由空间特征感知模块、时间运动特征感知模块和质量分数融合模块组成。为了提供与主观质量评价实验的平均意见分数(MOS)高度相关的质量分数，有效地利用和融合了FVV在空间和时间域的质量相关特征。实验结果表明，与现有方法相比，PLCC和SRCC分别提高了25.0%和18.5%。此外，烧蚀实验证明了空间特征和时间运动特征的重要性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE international Symposium on Broadband Multimedia Systems and Broadcasting

自引率

0.00%

发文量