> > Don't be too sure :) remember that ASL and it's (competitors? > equivelents?) are contextual and word/concept based, not spelling based. > Allthough you can fingerspell things, and do for stuff you don't know > the sign for, macros are sort of built in :) Still, I suspect that for > full input, that there would be the same type of issues and problems as > voice input. Signed words have "accents" as much as spoken words do. > People are just so darn good at pattern recognition. > In I think '92, I had Geoff Roberts, a master student of mine, implement a scheme I envisaged for recognition of gestures for Australian sign language. It incorporated not just Postures (finger positions), but sweeping hand movements and time-varying postures. It was done in association with the Victorian Deaf Society. Jim is correct, there is an awful lot of context in sign language including facial expression, body posture etc. It was clear from the beginning that without image input, the best we could achieve was a superset of finger spelling. BTW, it worked exceptionally well for what it was. Mike -- Subscription/unsubscription/info requests: send e-mail with subject of "subscribe", "unsubscribe", or "info" toWear-Hard Mailing List Archive (searchable): http://wearables.blu.org please, Please, *PLEASE* don't subscribe through a forward/false domain
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail