单摄像头条件下,基于桌面TUI,划分出4类具有不同场景语义的用户界面操作实物;分析实物用户界面中执行通用交互任务的特点,建立支持4类语义的3层系统设计框架;结合场景规划系统自身特征,设计了4类承载上述场景语义的标记,并提出标记任务分配策略;为解决大场景与小视域之间的矛盾,结合paddle技术,提出动态比例空间、动态成组、双手交互和遮挡处理等交互技术;借鉴认知心理学原理,提出时空复用交互技术;开发基于TUI的交互工具箱,并进行了应用验证.
Based on the notion of desktop TUI, four kinds of user interface elements are proposed on the condition of single camera. By investigating the main characteristics of common interaction tasks in TUI systems, we build a three-level system frame, supporting the four kinds of semantics above. Combined with the characteristics of the scene planner system itself, four kinds of markers are designed to carry the four kinds of semantics respectively; they are assigned to implement different interaction subtasks. In order to solve the contradiction between a wide scene and a narrow FOV, the techniques of dynamic scale space, dynamic binding, two-hand interaction and occlusion are proposed based on paddle. We also propose the interaction technique based on time and spatial multiplexing following the principles of cognitive psychology,. Furthermore, a toolkit based on TUI is built up, several applications based on these techniques are demonstrated.