Yes, it's emg+accelerometer detecting fingers position as hand gesture.
Although, it's difficult to call it a "gesture" recognition.
Besides, possible gestures seems to be limited to the recognition pattern they provide. Which overlook the real capability of gesture following, which is the possibility of defining gesture to recognise on the go.

As Pedro pointed there has been a good amount of research in this area, specially in HCI, for the past 5/6 years.
This is the natural outcome of those studies.
They're trying to bring to the general public and out of the labs a gestural biosensing device, and this is great.

Still, in my experience, emg sensing has very different calibration time, which depend on some bodily aspects unique to each person, and the environment where the user is. That is, for some, an emg sensor may calibrate in one minute, for others in 20mins.

They claim to have better (and yes, patented) electrodes.
Unfortunately, we'll see the device next year (claimed ship "late 2013").
As often these days, media run the (fast and soon forgotten) hype of a good 3d mockup and a well-edited video of applications.

Let's see what happens later on when we can try the real thing!
It's exciting to see a follow up in the industry to this specific artistic and research area of interest.
and I'm sure there will be more of those coming out.

from us too :P

--
Marco Donnarumma
New Media + Sonic Arts Practitioner, Performer, Teacher, Director.
Embodied Audio-Visual Interaction Research Team.
Department of Computing, Goldsmiths University of London
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Portfolio: http://marcodonnarumma.com
Research: http://res.marcodonnarumma.com
Director: http://www.liveperformersmeeting.net

On Sat, Mar 2, 2013 at 2:50 AM, <pd-list-request@iem.at> wrote:

Send Pd-list mailing list submissions to
pd-list@iem.at

EMG in action, detecting fingers through ML:
- http://www.youtube.com/watch?feature=player_embedded&v=6_7BzUED39A#!
(Scott Saponas, UIST 09)

On Sat, Mar 2, 2013 at 2:42 AM, Ivica Ico Bukvic <ico@vt.edu> wrote:

> EMG is potentially capable of this and with adequate filtering techniques
> there are papers out there that allow for fairly accurate detection of
> finger motions. More so, the video shows use of very distinct hand
> positions. In other words, it is not the gesture but hand shape that can be
> read accurately and interpreted. I suspect it also has a gyro/accelerometer
> that works in tandem with multiple EMG sensing points?****
>
> ** **
>