[FoRK] Annihilation from within...

Eugen Leitl < eugen at leitl.org > on > Thu Dec 7 09:17:42 PST 2006

On Thu, Dec 07, 2006 at 09:50:21AM -0600, Jeff Bone wrote:

> BTW, Eugen --- do you really think EEG's sufficient for 6DOF, much  

I think there's a lot of hidden signal covered under the noise in a
real multichannel EEG pickup (interestingly enough, the limit
are scalp electrodes -- EEG recorded directly from the brain
surface has lots more signal and much better resolution).
MEG doesn't have the interfacing problem.

> less low-bandwidth ASCII (assuming subvocalization?)  I'd been  

Using myopotentials (also dipole induction, as in EOG) 
wherever appropriate is definitely a great DSP saver. But anything
involving skin contact is going to be difficult. Degreasing with
isopropanol, a mild skin peeling and a gel will get you a long
way (whatever you measure you don't have to segment out by DSP), 
but this doesn't work at all well outside of the lab.

> assuming hybrid readers would be necessary.  If you think EEG's  
> sufficient for that, lots of interesting possibilities open up.

I suspect (based on no other evidence than current rate of progress) 
that one can get out a lot out of an EEG under optimal
conditions, using DSP to pick up signal from noise. Whether we will
be able to buy EEG controllers for Nintendo in ten years, no idea. 
 
> It's worth noting, I suppose, that IMHO you're going to need a whole  
> host of supplementary, "external" input signals for the interesting  
> apps:  at a minimum head tracking, probably eye tracking and pupil  

Since accelerometers in cosumer gear (Mac, Wii controller) have landed,
realtime headtracking is effectively solved but for lack of an appropriate
product one can buy.

> dilation measurement, skin conductivity, various kinds of  

Gaze tracking works well with video (NIR reflexion on cornea), 
or EOG even.

> proprioceptive localization inputs, ambient noise, and possibly extra  
> artificial visual inputs in various directions.  But none of that's  
> revolutionary.  The integrative capability --- by which I mean the  
> hardware and software to process all that --- is more challenging  
> than any one of those things...

We've been waiting for VR since mid-80s, and we still don't have it.
Wearables are still not there. Sometimes, things one thinks are solved
take forever in reality.

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE

More information about the FoRK mailing list