ネットワーク上の映像ストリーミングのためのインタラクティブシステム

我的研究是开发实时的人与人交互系统，使网络视频流媒体或通信更高效，不仅有利于网络流量，还能提高用户的感知质量。目前，用户仍然对网络流媒体感到不满，即使是更大的视频分辨率和更快的网络速度，因为我们是人，我们总是更喜欢更智能的视频感知体验。我们之前的工作隐含了观众与视频流服务器之间的交互，该服务器可以动态跟踪和预测观众未来观看的位置和区域，并通过优化的视频流媒体包进行反馈。根据今年的工作安排，我们在主观实验方面进行了更多的研究，主观实验分数真实地反映了交互系统在视频流媒体上的准确性。我们购买了更可靠的计算机和网络连接存储设备来采集和存储我们的主观实验数据，并使用了更精确的眼动跟踪设备进行实验。为了改进我们的主观实验，我们通过会议或访问的方式与教授和研究人员进行了会议，以确认协议和标准。为了使结果更准确，我们邀请更多的人加入我们，详细信息可以在下面提供的URL中找到。数据来源：http://http://klab.nii.ac.jp/~fengdragon/jsps_proj/

My research is to develop real time Human Interactive System to make network video streaming, or communications more efficient, not only to benefit the network traffic, but also to improve users’ perceptual quality. For now, users are still feeling discontented over network streaming, even with larger video resolutions and faster network speed, because we are human, and we always preferring for more intelligent video perception experience. Our previous work has implied interactions between viewers and video streaming server, which could track and predict viewer’s future gaze location and area dynamically, and then feedback with optimized video streaming package.As scheduled for this year’s work, we investigate more on subjective experiment, for the subjective experiment scores are truly reflecting the accuracy of interactive system on video streaming. We purchased more reliable computer and network attached storage devices to capture and store our subjective experiment datas, also much accurate eye tracking equipment was used for the experiment. Also the storagesystem is used for a database source website, where we publish our raw data and videos for other researchers.To improve our subjective experiment, we had meetings with Professors and Researchers via conference or visiting to confirm the protocol and standard. And to make the result more accurate, we invite much more people to join us, the detailed information could be found at the URL provided below.DataSource: http://http://klab.nii.ac.jp/~fengdragon/jsps_proj/