On the Environment of the Virtual Real-Time Media via the Networks

本研究主要研究了在网络环境中具有传输滞后的视觉和触觉信息的呈现方法，虚拟环境中立体镜的调节和收敛对视觉和触觉信息的影响，实时交互与网络滞后的关系，以及4）针对由滞后引起的劣化的预测方法的交互式应用的示例。在第1）项中，通过使用视觉和触觉显示来考虑实时交互环境以在视觉信息之前基于触觉信息来开发协议，因为前者在传输性能方面具有益处，并确认该方案的优点。其次，在2）中，通过任务性能的实验，通过使用这两个参数可控的HMD，确认收敛和调节之间的一致性有助于性能。在3）中，考察了具有滞后的网络上远程用户之间的协同工作的任务性能，其中评估了远程用户的预测的实现。最后在4）中，作者尝试将基于预测的交互方案应用于网络用户的竞争环境，即在线赛车游戏。引入新型的预测模型以减少不兼容性和不舒适的交互，其中每个终端协同地在每帧中修改预测模型以减小每个终端中预测的相对化身与本地化身之间的距离的差距。未来的工作包括对该预测模型的分析以及在其他背景下的应用。

In this research, the authors study on1) the method of presenting visual and tactile information on the network environment with transmission lag,2) the effect of the accommodation and the convergence for the stereo scope in the virtual environment,3) the relationship between the real-time interaction and the lag on the network, and4) the examples of interactive application of the prediction method against degradation caused by the lag.In the item 1), the real-time interactive environment is considered by using visual and tactile display to develop the protocol based on tactile information prior to visual information since the former has the benefit in terms of transmission performance, and to confirm the advantages of the proposal.Secondly in 2), through the experiment of the task performance, it is confirmed that the consistency between the convergence and the accommodation contributes to the performance, by using the HMD of which those two parameters is controllable.In the item 3), the task performance of collaborative work between the remote users on the network with lag is examined, where the implementation of prediction of the remote user is evaluated. As a result, forty through fifty percent of performance improvement is confirmed even if the simplest prediction model is utilized.Finally in 4), the authors try to apply the prediction based interaction scheme for the competitive context of the network users, namely the online racing game. The new type of prediction model is introduced to reduce the incompatibility and comfortless interaction, where each terminal, in collaboration, modifies the prediction model in every frame to reduce the gap of the distance between the predicted opposite avatar and the local avatar, in each terminal. Future works involve the analysis of this prediction model and the application for other context.