当前位置:
X-MOL 学术
›
arXiv.cs.MM
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
Spatial Privacy-aware VR streaming
arXiv - CS - Multimedia Pub Date : 2021-04-29 , DOI: arxiv-2104.14170 Xing Wei, Chenyang Yang
arXiv - CS - Multimedia Pub Date : 2021-04-29 , DOI: arxiv-2104.14170 Xing Wei, Chenyang Yang
Proactive tile-based virtual reality (VR) video streaming employs the current
tracking data of a user to predict future requested tiles, then renders and
delivers the predicted tiles to be requested before playback. Very recently,
privacy protection in VR video streaming starts to raise concerns. However,
existing privacy protection may fail even with federated learning at head
mounted display (HMD). This is because when the HMD requests the predicted
requested tiles and the prediction is accurate, the real requested tiles and
corresponding user behavior-related data can still be recovered at multi-access
edge computing server. In this paper, we consider how to protect privacy even
with accurate predictors and investigate the impact of privacy requirement on
the quality of experience (QoE). To this end, we first add extra camouflaged
tile requests in addition to real tile requests and model the privacy
requirement as the spatial degree of privacy (sDoP). By ensuring sDoP, the real
tile requests can be hidden and privacy can be protected. Then, we jointly
optimize the durations for prediction, computing, and transmitting, aimed at
maximizing the privacy-aware QoE given arbitrary predictor and configured
resources. From the obtained optimal closed-form solution, we find that the
increase of sDoP improves the capability of communication and computing hence
improves QoE, but degrades the prediction performance hence degrades the QoE.
The overall impact depends on which factor dominates the QoE. Simulation with
two predictors on a real dataset verifies the analysis and shows that the
overall impact of sDoP is to improve the QoE.
中文翻译:
感知空间隐私的VR流
基于主动图块的虚拟现实(VR)视频流使用用户的当前跟踪数据来预测将来请求的图块,然后在回放之前渲染并传递要请求的预测图块。最近,VR视频流中的隐私保护开始引起人们的关注。但是,即使在头戴式显示器(HMD)上进行联合学习,现有的隐私保护也可能会失败。这是因为,当HMD请求预测的请求切片并且预测准确时,仍可以在多访问边缘计算服务器上恢复实际的请求切片和与用户行为相关的数据。在本文中,我们考虑即使使用准确的预测变量也可以保护隐私,并研究隐私要求对体验质量(QoE)的影响。为此,我们首先在真实图块请求之外添加额外的伪装图块请求,然后将隐私要求建模为隐私的空间度(sDoP)。通过确保sDoP,可以隐藏实际的图块请求,并可以保护隐私。然后,我们共同优化了预测,计算和传输的持续时间,目的是在给定任意预测变量和配置资源的情况下最大化隐私感知QoE。从获得的最优封闭形式解中,我们发现sDoP的增加提高了通信和计算的能力,从而提高了QoE,但降低了预测性能,从而降低了QoE。总体影响取决于主导QoE的因素。在真实数据集上使用两个预测变量进行的仿真验证了该分析,并表明sDoP的总体影响是改善QoE。
更新日期:2021-04-30
中文翻译:
感知空间隐私的VR流
基于主动图块的虚拟现实(VR)视频流使用用户的当前跟踪数据来预测将来请求的图块,然后在回放之前渲染并传递要请求的预测图块。最近,VR视频流中的隐私保护开始引起人们的关注。但是,即使在头戴式显示器(HMD)上进行联合学习,现有的隐私保护也可能会失败。这是因为,当HMD请求预测的请求切片并且预测准确时,仍可以在多访问边缘计算服务器上恢复实际的请求切片和与用户行为相关的数据。在本文中,我们考虑即使使用准确的预测变量也可以保护隐私,并研究隐私要求对体验质量(QoE)的影响。为此,我们首先在真实图块请求之外添加额外的伪装图块请求,然后将隐私要求建模为隐私的空间度(sDoP)。通过确保sDoP,可以隐藏实际的图块请求,并可以保护隐私。然后,我们共同优化了预测,计算和传输的持续时间,目的是在给定任意预测变量和配置资源的情况下最大化隐私感知QoE。从获得的最优封闭形式解中,我们发现sDoP的增加提高了通信和计算的能力,从而提高了QoE,但降低了预测性能,从而降低了QoE。总体影响取决于主导QoE的因素。在真实数据集上使用两个预测变量进行的仿真验证了该分析,并表明sDoP的总体影响是改善QoE。