Slashdot Mirror
A No-Touching 3D Computer Interface
Justin Schunick points out a video demonstration of a 3D input system which senses the user's hand position, but without requiring the user to touch a controller or wear a trackable position indicator. From the provided description: "Utilizing the theory of electrostatics, we have designed a low-cost human-computer interface device that has the ability to track the position of a user's hand in three dimensions. Physical contact is not required and the user does not need to hold a controller or attach markers to their body. To control the device, the user simply waves their hand above it in the air."
Why? a little counter-intuitive, my tablet can do that much better.
I think it's an awesome idea, but poorly executed examples.
"...the user simply waves their hand above it in the air..." These are not the droids you're looking for.
Good, I didn't want all those damn dirty, germ ridden, greasy fingerprints all over my nice shiny monitor.
Why save your soul when you can sell it for a profit?
... then maybe a no-touch interface is not going to work well for me. I found the Theremin to be almost impossible to play because there was no way to get my hand in exactly the same x-y-z coordinate and with the same roll-yaw-pitch attitude (all of which affected the frequency of the oscillator). YMMV, of course.
Interesting from a tech, nerd perspective I suppose. However, a web cam and a computer vision gesture control app can produce the same effect much more efficiently.
Your arms are going to get tired very quickly using this interface... Maybe we should rather work on perfecting those mind control interfaces.
A picture is worth exactly 1024 words.
From The Hitchhiker's Guide to the Galaxy, by Douglas Adams.
When our name is on the back of your car, we're behind you all the way!
A lot of these technologies are really waiting more for computing power to increase to a level where it can support it comfortable, more then new ideas on how to get it to work.
We know how to take 2 camera and generate a 3d model of what the cameras see. The problem is processing speed.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Just needed to turn down the sensitivity a little... like people who put their mouse on super slow tracking vs. those who prefer a quick twitch approach. Personally I'd like a trackpad replacement with this interface, I don't want to have to move my whole arm around. Make it plenty sensitive or rather make the sensor's grid scaled appropriately for the size of the input.
A fool throws a stone into a well and a thousand sages can not remove it.
Except with electrostatics instead of heterodyning?
I don't know about everyone else, but holding my hands anywhere in free space takes quite a bit of energy unless they are hanging at my sides. The reason the keyboard and mouse or other touch surfaces work well is because they allow a person to rest their limbs in an unnatural position.
So I am not sure about anything that doesn't allow a person to rest... it'd be like using a whiteboard all day long, and that is quite tiring!
This post brought to you by your friendly neighborhood MBA.
If I were using this on a desktop, I wouldn't mind wearing a small button on my hand to allow me to click easier (squeezing your thumb and index finger is less effort than moving your whole arm forward) and maybe have a small brace to rest my wrist on, one that gimbals around, to save my arm from being tired.
If it could be made simpler and integrated with mobile devices I could see it begin a winner though. Tiny mice and track pads are horrible, Touch screens have always been my prefered mobile input device and one that lets me use the computer with my fingers without smudging the screen would make me happy :D
Do I sense a Theramin app for the iPhone 5G?
It pays to be obvious, especially if you have a reputation for being subtle.
Who else read this headline and immediately thought of Arrested Development?
"No touching!"
Mod Up!
Make it a hybrid interface.
Big touchpad where the mouse is now, plus the ability to recognize gestures above it.
Monitors are for looking at.
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
Theramins suk.
Make this a finger gesture interface. Your wrist is resting, and your fingers can do stuff fairly repeatable.
(Reboot from BSOD = That Gesture.)
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
First I want to see what is actually under the cloth, second I want to know if he is just using a multiple pickup therimin, one for each axis.
I balked at the cloth as well, but stuck out the video to the last quarter, where they pulled it back and gave a short tour of the guts of the system.
Looks pretty simple. I don't think any new discoveries have been made, but rather the device is a clever bit of engineering using known electrical properties. Applied and scaled correctly, it holds some neat possibilities, though without some form of tactile feedback it might be awkward to manage information with one's hands in that manner. But who knows? Humans are good at adapting. The basic keyboard input seemed pretty impossible when I was a kid, but now my fingers are able to fly across the keys.
-FL
Actually, I think a full-blown sign-language would be great, especially for use with cell phone cameras. There are a few obvious benefits, and some not-so-obvious ones:
(1) people learn international sign language, and it assists in international communication.
(2) The speed of data entry would be increased greatly.
(3) It seems to me probable that there would be decreased cost and possibly (if it was done by a designed/dedicated chip) decreased battery usage by using sign lanugage instead of other means
(4) Logon would be simplified, with simply flashing a thumbprint.
How to do it? I suspect that the way to do it might be as follows:
(1) take an image, and subtract one image's RGB map from the previous image.
(2) Run an FFT on the result, to get a motion map.
(3) Track the motion of the various blocks of pixels.
(4) From the motion of the various blocks, and moreover from what remains invariant and what adds on on one side (or disappears from the other), obtain a 3-D map of various objects. From the 3-D map, and how it morphs, obtain approximate rotation vectors.
(5) Recognize hands by the digit lengths and connection combinations.
(6) Plot the hand digit rotational and bending angles into a real-time motion map.
(7) Translate #6 into specific signs, which in turn can be programmed to be equivalent to international sign language.
The above method would also allow very high levels of compression of video.
Correct Horse Battery Staple: 72 bits of entropy. Enter "Correct H" into google. When it generates the phrase, that's
(1) people learn international sign language, and it assists in international communication.
I'm a Hungarian, living in the UK, posting on an American website. Don't tell me about international communication.
(2) The speed of data entry would be increased greatly.
Compared to a touchscreen, maybe. Assuming of course the software can translate SL into your native language. Compared to a keyboard, this is a joke. Try coding C in sign language, and report back when they let you out from the mental institute.
(3) It seems to me probable that there would be decreased cost and possibly (if it was done by a designed/dedicated chip) decreased battery usage by using sign lanugage instead of other means
You mean a high quality (especially considering #4) camera and a custom-designed and -manufactured chip or recognition software with the associated CPU load vs. a $10 keyboard?
(4) Logon would be simplified, with simply flashing a thumbprint.
Fingerprint is not a secret. Repeat that until it sinks in. (They use it to catch criminals because we leave them all over the place, you know.)