Keynote Lecture - Perceptual Interfaces
Professor David Hogg, University of Leeds
Abstract
It is widely believed that computers will be easier to use if we can
communicate with them in ways that are more similar to our interactions
with other people. To achieve this will require advances in animation
(e.g. facial modelling), perceptual technologies (e.g. computer vision),
and cognitive aspects of interaction. A major research challenge is to
find ways for acquiring and encoding the wide spatial, temporal and
procedural variations between and within different types of interaction.
We describe recent work on visual gesture recognition and interaction
modelling in which the range of possible things that can happen are
learnt automatically through passive observation of video sequences
depicting typical gestures and interactions. The basis for the approach
is the construction of probabilistic spatio-temporal models from
training data extracted from the video sequences.