When people collaborate they often use gestures as an important part of their communication.  Gestures can take many forms, but people often use a form of gesture called deixis – an indicative gesture that commonly takes the form of pointing – to identify objects, locations, or directions.  This behaviour is particularly critical for communication when people are collaborating over a shared artifact, such as a map on a table. Map-based collaboration, or geocollaboration, occurs in many domains, including urban planning, emergency response, foreign and domestic policy, and a variety of research and industrial settings. In most of these settings, the collaboration can be described as surface-based (i.e., it occurs over a table or whiteboard).

Although collaborations are often collocated, they are increasingly distributed, taking place in multiple locations, at different times, or both. In distributed, surface-based geocollaborations, communication is usually improved through the use of embodiments: representations of remote users, their characteristics, and their activities. Although embodiments can show some deixis, current implementations are limited and as a result, distributed collaboration over shared artifacts requires more time, is less expressive, less successful, and more error-prone than collocated collaboration. In some domains, such as emergency management and political negotiation, such problems can mean the difference between success or failure.

This seminar presents my research into how to address these problems, including ethnographic analysis of how gestures are used, the design and evaluation of embodiment candidates for representing gestures in distributed settings, and the development of a toolkit, KinectArms, for capturing and showing gestures over distributed surfaces.