Rapidly Deployable Displays for Emergency Response

Rapidly Deployable Displays for Emergency Response Christopher Jaynes, Ruigang Yang October, 2004 Abstract This project will focus on the scientific and engineering problems related to ultra-portable, rapidly deployable visualization systems (RDVS) for emergency response. High-end visualization systems are simply too rigid, costly, and cumbersome to support first responders in the field who are faced with difficult command and control, decision making, and planning problems. Traditional visualization systems require a large footprint and support infrastructure that cannot be present in situations that require visualization on-demand. This project will build on our past success in multi-projector, self-configuring systems and work towards ultra-portable immersive display systems that support high-resolution visualization for emergency response scenarios. The project will focus on two major problems; 1) High-resolution display anywhere using commodity projector arrays, and 2) Novel user interface mechanisms that utilize cameras to track user position/gestures within the display. The researchers will deliver a prototype display system that is portable, reconfigurable, and self-calibrating. Because the display will be composed of a cluster of commodity projectors and computers, the display can be used to support different visualization senarios from largeformat video playback, to immersive four-wall surround simulation and training. Components are low-cost and lightweight and can be carried by responders in the field. The display will be capable of delivering 11 million pixels of resolution with a setup time of less than 20 minutes. In addition, the research project will generate new results in the area of gesture recognition and tracking. An initial portable display system will be developed and demonstrated in year one. In the second year of the project more sophisticated technique for gesture tracking and analysis and new approaches to color and geometrically correct projection on arbitrary surfaces will be integrated with the display prototype. 1 State of the Art /Practice Multi-projector, visually immersive displays have emerged as an important tool for a number of applications including scientific visualization, augmented reality, advanced teleconferencing, and simulation. Unfortunately, high-resolution visualization environments are only found in a few facilities and research institutions. This is largely due to the high costs associated with both the initial purchase and the long-term maintenance of these systems. The advent of consumer-grade projectors and advances in cluster-based rendering promises to reduce the overall system cost, although system maintenance costs remain very high. These costs are the result of the fixed mechanical infrastructure and specialized hardware that are normally associated with immersive displays. Specialized hardware components provide speed and parallelism through the use of large shared memory machines and expensive multi-headed output. Use of these immersive systems requires significant re-engineering of the environment to accommodate the system. In addition, special display surfaces must be installed and projectors are constrained to prespecified configurations. Once these displays have been installed, they must be carefully calibrated by-hand. If projectors must be moved or replaced, a time-consuming re-calibration process must be undertaken. Fixed installation displays cannot be scaled or reconfigured according to the users' changing needs and are prohibitively expensive for widespread use. These observations have provided the impetus for our research program that seeks to design, implement and deploy the immersive display environments that are composed of low cost, commodity