[oats-sig] new member of mailing list

I really like the idea of the gaze-driven switch. I also, really, really,

Hi All

I did some work on this over the weekend, mostly building and setting
up opengazer and playing around with some ideas. Its pretty fun once
you get it working.

The next step is to start interpreting the calculated gaze positions
into a switch activation. opengazer outputs everything to a port as it
works, which should make things a bit easier. Any suggestions on good
assistive software I can play around with? I hear there's this great
"Jambu" thing... :)

I'll probably start with a simple threshold distance to a corner of
the screen and see how that works out. The accuracy is terrible with
this webcam, but it should do for a basic switch.

What would be really great is if we could get opengazer running with
other webcams to compare accuracy.

Any takers? ;) I can put up some instructions on how to build the
whole thing. I should probably warn in advance that it is a bit of a
pain to get right because vxl needs to be built from scratch, but well
worth it! If anyone is willing to lend a hand I think building some
debs would really help take the pain out of it.

On 9/28/07, Simon Judge <simon.judge at nhs.net> wrote:
> I really like the idea of the gaze-driven switch.  I also, really, really,
> like the idea of a gesture-driven switch (i.e. choose the movement, system
> learns, repeat the movement, system switches)...

I really like that idea! It sounds like a similar approach to Andrew
Duchowski's Eye Writing that he showed at COGAIN.

Perhaps a better camera (though still in the cheap off-the-shelf
range) could deliver a good enough accuracy to achieve something like
this, though I dont know for sure that camera resolution is the
limiting factor with opengazer. I would be useful to get Piotr's input
on this.

> I look forward to seeing the first betas Stephen! (-;

me too. Now all I need is some free time!

Stephen Emslie