A Japanese research group has developed a Laser-based UI to facilitate Human-Computer interaction. The device is meant to minimize the need for extra space on small devices. The input is made via a laser that tracks a finger in real-time and 3D. the user then navagates through a custom interface either on a normal computer or even via small portable devices.
Given what we know about Flash 8, this seems overly complex. Heck, given what the playdoCAM creator told us, this was possible via Director MX 2004. It should be possible to create a similar interface in Flash 8 using the Camera object and the new bitmap features. Then you devices could use this new technology via a dual use part. Now your camera phone doesn't need a laser attachment because it has a camera that can do almost the same thing. Now I'm not saying this isn't a big accomplishment, I'm just saying we should try and do the same thing via existing technology just to find out if it can be accomplished cheaper.
via MAKE
Latest Katrina update: Times Picayune cartoon thanks Texas
1. sampling the webcam graphic allows you to determine the colour of a given pixel, and sampling per frame can tell you if a colour has changed. This is not the same as using a laser.
You would need to calibrate the webcam to determine the shades of skin tone which could work, but fingers and hands hovering would also be picked up. Also sampling the area covered by a keyboard often enough would be impossible with all but the very fastest pc, even if you are dealing with a 240*360 web cam at 12 frames a second thats 1036800 calculations a second!
I really dont think this would be possible, i'm pretty certain it wouldnt be possible to get even the simplest word out.
The most accurate thing ive seen so far is a barcode scanner which was pretty impressive, but it would be a very big jump to go any further than that.
I have been playing with motion capture with flash 8 and managed to get basic hand/body detection working.
Heres more technical info about how tricky it is.
http://ilab.cs.ucsb.edu/projects/mathias/KolschTurk2004RobustHandDetection.pdf
Saying all that - would love it if i was proved wrong.
hmm perhaps if you applied bright orange nail varnish?
Posted at 2:39PM on Sep 7th 2005 by mh