Work
Package 4: Animation & Modelling
This
is the first of the technology work packages underlying
the planned applications described above. It builds upon
the experience of virtual human avatars which will soon
be commonplace in all communication environments. Individual
components of shape and movement may be acquired from life
(motion capture of human signers) or constructed in 3D graphic
space using physical modelling tools. Tools based on each
system will be developed in parallel, evaluated, and deployed
as appropriate. For use with the direct recording of signed
sequences, the project will develop a refined suite of advanced
Motion Capture Tools, forming a single, coherent, capture,
recording and replay system, using robust techniques and
equipment capable of use in TV studios and other industrial,
non-laboratory settings.
|
|
Work
Package 5: Language and Notation
This
is the second of the underlying technology packages. This
package will move from the presentation of manually encoded
versions of sign language (such as Sign Supported English)
generated from text - as in the 'Simon-the-Signer' prototype
system previously developed by ViSiCAST participants - to
the much more complex problem of describing, encoding, and
presenting natural European sign languages, derived from
English textual sources. The enabling technology here is
a sign language notation already under development by the
University of Hamburg, HamNoSys, which will be further developed
to give an XML-compatible notation, ViSiCAST-GML. The required
lexicons will be developed using this notation.
|