Fraunhofer’s FIT . . . has recently appeared on the YouTubes, where we must say it looks pretty darn good. Not only does it not require special gloves or markers, this thing also works in real time and can support multiple users (and multiple fingers).
This is way cool. Although what is really needed is some touch (or touch-free) everyday interface beyond pictures
first… no lag??? lag is terrible just as is recognition frame rate, which is downright catastrophic.
second???? that kinda interface (minority report like) might be interesting and cool for 5 minutes, later it would just became painful and tiresome. Holding hands in the air for few hours is not something anyone would want. looks cool, but definitely not usable in real life. only really usable feature of such controlling is full motion body capture for games and special effects.
don’t get me wrong, i’m all for motion detection, as much as i’m interested for correct approach.
well I don’t see it so negative, it’s a prototype and from what it does you can easily imagine a room covered with projectors and recording everything you do, using a computer becomes just the same as using a room, you touch a door projected onto your wall and enter a 360° game where you actually move around. sure you’d need a higher recognition rate but that will come as sure as the processing units that can calculate a realistic 360° environment. make it a small room for games with a moveable floor and you’re basically all set for your typical fps…
Let’s just hope nobody deactivates the safety protocols otherwise we’ll have a standard holodeck killing.
lol
to more serious note, you’d only need something like dualshock on various places of your body, vibrations make wonders on controllers. although… vibrations comming from hitting someones crutch might cause completely different effect;)
lol… kinda too expensive with projectors
simple 3d glasses would suffice beside the point you’d really be in that world. but… more to the problem, there is a big downside on body motion control in games. absolutely no feedback. let’s say you play action adventure and monster smacked you on the ground… now your avatar and you reside in completely different positions. or let’s say boxing, your avatar gets smacked and loses balance while you didn’t. in that case the only obvious choice for developers is predictive movement which might make game feeling right for viewers, but your char will practically enter video sequence until getting into your position with no choice of other thing. there were a lot of studies on that topic already.
as far as motion tracking in games goes i still think wii hit the nail and Move improved that same tech to the max. simple, but intuitive and most importantly accurate, which 3d body tracking isn’t
http://www.physorg.com/news168797748.html
Nah for a boxing game the player just need a strong taser. When you get knocked to the ground the taser will just tase you to the ground =)
no but you are right no feedback really sucks being able to hit right though every opponent wont be any fun. Then that would leave two options every opponent gets thrashed when you hit them, but then the game would not be so fun since you would be super strong. The other option is that what you see and what you feel will be a contradiction which sucks.
We would have to interface nerves to be able to lock up muscles then we are starting to talk. I would not like to test the prototype for that system
Uh… no thanks, what if a bug locks up my hearth?
I prefer a system that injects sensations into your nervous system but does not block anything. It’s enough to be tied to a couch to prevent muscle movement, after all the movement it’s just a sensation.
And of course, a triple redundant system that brings you back into reality if something is wrong.
Wonder what Kinect will bring ..
Exactly… nothing.
Original kinect was first offered to nintendo… they declined
Then it was offered to Sony… they declined.
(There are videos of kinect usage on youtube done by Sony on ps2)
Ms bought them on the other hand simply because they lacked motion controller and to make matters even worse they completely crippled hardware by removing all detection parts to lower its costs. Now… do you think that Nintendo and Sony don’t know what they do?
Problem is insane resolution 3d camera would need to detect you well. Kinect doesn’t detect fingers, just body parts
Even if this looks very promising, the whole demo relies on a basic 3-d model. There are many multi-touch demos showing the manipulation (moving/zooming) of images, but demonstration of real usable applications are rare.
It would be nice to see how this technique could be useful to replace the mouse-clicks-ok-button on a normal desktop. Perhaps replacing the touch functionality that is becoming popular on todays big computer screens.
When programming, moving the hand to the mouse to move the cursor is often time-consuming. It would be much faster to just raise your hands and make a gesture to the screen.
Regards,
Claes
“Not only does it not require special gloves or markers”
but one would like to know if it requiters a room with solid wall colors and not to colorful cloths. Looking at the video i kind of gets the feeling that the settings in the room are heavily controlled.
would the system really work in a normal bright room and for a person wearing bright clothes
GestureComputing !!!