A scaleable sensor approach for immersive and desktop VR applications

All modern VR and AR toolkits provide some kind of sensor abstraction for different types of input and output devices and user interaction. Almost all systems today define a single layer of abstraction. Most toolkits usually support only low level data streams leaving most of the user interface implementation work to the IT professional. Some provide only high level interaction responder (e.g. X3D) without the possibility to access low level I/O information and streams. No toolkit supports adoption of UI elements to the environment the user is working in, e.g. immersive or desktop. In our approach we build three level of abstraction for user interaction which leads to a more flexible and scalable result. Application programmers are free to use low level streams (e.g. tracker data) directly, high level responder (e.g. the user `touched' something) and even application requests (e.g. the systems needs a float input or a list of selected objects) which are automatically mapped to 2D UI elements or immersive responder, depending on the environment and used devices. Two examples illustrate the flexibility and efficiency of our approach.