CSR-DMSS, SM: View Control Management in Geographically Distributed Tele-Immersive Environments

CSR-DMSS、SM：地理分布式远程沉浸式环境中的视图控制管理

• 批准号: 0834480
• 负责人: Klara Nahrstedt
• 金额: $ 42万
• 依托单位: University of Illinois at Urbana-Champaign
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-09-01 至 2012-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0834480&HistoricalAwards=false
• 关键词: CSR; DMSS; SM; View; Control

Revised Budget/StatementAbstract for our project (about 250 words) Emerging 3D Geographically Distributed Tele-immersive Environments (3DTI) are becoming viable collaborative environments over the Internet to allow for joint physical activities such as tai-chi training and many other usages. These environments fuse multi-dimensional streams from multiple 3D cameras and other sensors across multiple sites to render a shared multi-view visual cyberspace at each site that immerses all participants. In the cyberspace, each participant manipulates herself (e.g., positions, size), and interacts with remote participants or graphical objects ?virtually?. A fundamental challenge in these environments is the need for effective real-time interaction among multi-site geographically distributed 3DTI environments and the PI observes that maintaining a consistent, information-rich shared visual context is the key to effective real-time interaction. If the views of different participants become out of sync, their collaboration is severely disrupted. In this project, the PI presents a novel view control management, called VIEWCALL. VIEWCALL presents (1) a holistic approach to view dissemination where the PI investigates forest construction and load balancing for view-centric stream dissemination among individual 3DTI environments; (2) a novel approach to visual context/feedback distribution due to dynamics in view or resource demands, and (3) new distributed algorithms for capturing, updating, coordinating and synchronizing view-based visual context metadata. The VIEWCALL system has an impact on the 3DTI environments where research community investigates new forms of dance choreography, students learning tai-chi, in-home care of ageing population, computer gaming, and wheelchair-basketball coaches training players with movement disabilities. VIEWCALL is regularly shown at the annual UIUC?s Engineering Open House as well as to ChicTech all-schoolgirl summer camp.

修订的预算/声明我们的项目摘要（约250字）新兴的3D地理分布式远程沉浸式环境（3DTI）正在成为可行的协作环境，通过互联网，以允许联合身体活动，如太极训练和许多其他用途。这些环境融合了来自多个站点的多个3D摄像机和其他传感器的多维流，以在每个站点呈现共享的多视图视觉网络空间，使所有参与者沉浸其中。在网络空间中，每个参与者都操纵自己（例如，位置，大小），并与远程参与者或图形对象交互？实际上？在这些环境中的一个根本挑战是需要有效的实时交互多站点地理分布的3DTI环境和PI观察到，保持一致的，信息丰富的共享视觉环境是有效的实时交互的关键。 如果不同参与者的观点变得不同步，他们的合作就会严重中断。在这个项目中，PI提出了一种新的视图控制管理，称为VIEWCALL。VIEWCALL提出了（1）视图传播的整体方法，其中PI研究了各个3DTI环境中以视图为中心的流传播的森林构建和负载平衡;（2）由于视图或资源需求的动态性，视觉上下文/反馈分布的新方法，以及（3）用于捕获，更新，协调和同步基于视图的视觉上下文元数据的新分布式算法。VIEWCALL系统对3DTI环境产生了影响，研究社区调查了新形式的舞蹈编排，学生学习太极拳，老年人的家庭护理，电脑游戏和轮椅篮球教练训练运动残疾球员。VIEWCALL定期在一年一度的UIUC？的工程开放日以及芝加哥理工学院的所有女生夏令营。