A multi-projector CAVE system with commodity hardware and gesture-based interaction
Spatially-immersive systems such as CAVEs provide users with surrounding worlds by projecting 3D models on multiple screens around the viewer. Compared to alternative immersive systems such as HMDs, CAVE systems are a powerful tool for collaborative inspection of virtual environments due to better u...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | informe técnico |
| Fecha de publicación: | 2017 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/99668 |
| Acceso en línea: | https://hdl.handle.net/2117/99668 |
| Access Level: | acceso abierto |
| Palabra clave: | Virtual reality Virtual reality systems Immersive systems Projector calibration Gesture-based interaction Realitat virtual Àrees temàtiques de la UPC::Informàtica::Sistemes d'informació::Interacció home-màquina |
| Sumario: | Spatially-immersive systems such as CAVEs provide users with surrounding worlds by projecting 3D models on multiple screens around the viewer. Compared to alternative immersive systems such as HMDs, CAVE systems are a powerful tool for collaborative inspection of virtual environments due to better use of peripheral vision, less sensitivity to tracking errors, and higher communication possibilities among users. Unfortunately, traditional CAVE setups require sophisticated equipment including stereo-ready projectors and tracking systems with high acquisition and maintenance costs. In this paper we present the design and construction of a passive-stereo, four-wall CAVE system based on commodity hardware. Our system works with any mix of a wide range of projector models that can be replaced independently at any time, and achieves high resolution and brightness at a minimum cost. The key ingredients of our CAVE are a self-calibration approach that guarantees continuity across the screen, as well as a gesture-based interaction approach based on a clever combination of skeletal data from multiple Kinect sensors. |
|---|