Matching Technology to People for Telepresence

Abstract

The first steps towards a multi-media telepresence revolution have already been taken. Ubiquitous network connectivity with variable bandwidth on demand will be available in the near future due to the evolution of Global IP-networks and fixed mobile convergence, e.g. UMTS. In order to efficiently deliver the radical variety of services that will become possible it is necessary to match the delivery technology to the end user’s needs. It will be necessary to enable a range of communications tasks with a variety of bandwidths, local processing power, and user requirements. To do so the delivery systems will have to be adaptive and context sensitive. The network, middleware and delivery technology must be optimised using perceptual criteria. These criteria will be in terms of audio/video perception and communications task requirements. Perceptual models are being developed at BT Labs. For multi-media systems we require multi-modal models for which the influence of task is very significant. In order to account for this level of task dependency we are having to develop and evaluate cognitive models. Examples that illustrate the diversity of emerging service propositions and the complexity of the optimisation task they represent include: immersive telepresence environments, desktop video and video-graphic conferencing tools, portable multi-media terminal, and wearables.