Wiimote Turns TV into Touchless MS Surface

RemyBR writes "User interface project allows you to control objects on a display using gestures, working like Microsoft's Surface but without touching the screen at all. Inspired by Johnny Chung Lee's work, the system requires you to wear Minority Report-style gloves equipped with infrared emitters on your fingertips. A Wiimote on top of the display keeps track of these IR LEDs, while the software can read the motion down to two-finger pinching gestures for image zooming."

What style gloves?

[techpawn (969834)]

the system requires you to wear Minority Report-style gloves

come on! Don't toy with my emotions!!! Power glove man! System requires you to wear modified power gloves!

Re:

[AKAImBatman (238306)]

I love the Power Glove. It's so baaad.

Re:

[morgan_greywolf (835522)]

I love the Power Glove. It's so baaad.

Oh, c'mon. This one's on YouTube for cryin' outloud. At least provide some linkage [youtube.com].

Re:What style gloves?

[AKAImBatman (238306)]

If you actually need a YouTube video to know what I'm talking about, you need to hand in your geek card. NOW.

Re:What style gloves?

[AKAImBatman (238306)]

No, you lose (sorry, loose) points for that. Tom Cruise stars in that movie, remember? Which means you get scored -1 Scientology Wacko Supporter [youtube.com]. :-P

Gloves + Compiz-Fusion

[vivin (671928)]

When I saw this first on Johnny Chung Lee's site, I had the same idea (gloves et al). I was thinking how cool it would be if this could be combined with Compiz-Fusion or something like that (imagine manipulating the cube with your gloves!). I wouldn't mind trying to write something like this (but I don't have very much knowledge about writing code for X so not really sure where to start at all). I think it would be really neat. I was thinking that you could use gestures to manipulate the desktop. Either th

Bring Back the Power Glove!

[StCredZero (169093)]

For my undergrad digital hardware project, I made one of these act as a Macintosh mouse. That was cool my professor let me get away with such an easy project.

http://en.wikipedia.org/wiki/Power_Glove [wikipedia.org]

You could have a very large vocabulary of gestures by using finger positions like modifier keys. Apparently, the native resolution of the Power Glove is 8 bit. This might not sound like much, but with a smoothing function like the one used for SmartNav head pointing devices, you can emulate much higher resol

Stop the Wii shipments!!!

[kellyb9 (954229)]

Nintendo turns your televisions into a Microsoft product!

Table

[pavon (30274)]

Would it be possible to shine IR light through the edge of a plexiglass surface, and then when the user touches the surface it would cause the IR to scatter at that point creating a point source for the Wiimote to track?

Re:

[Aladrin (926209)]

That's crazy. Just crazy enough to work.

Actually, I think it sounds like a great idea... I wish I had the parts and time to try things like that.

Potential issue: Dust, fingerprints, etc... Wouldn't they also cause the same effect?

Maybe there's a material that causes IR light (or even all light) to reflect (or reflect differently) in an area where force is applied?

Re:

[slobarnuts (666254)]

Yea, pretty much the same theory behind most of the homemade FTIR boxes (google it, or check out the NUIgroup), instead of using a webcam with IR filter removed, just use a wiimote. The system doesn't provide the best tracking though.

Re:

[dustbowl (809387)]

You can even do it with a regular webcam and some IR lights - check out the forums at http://nuigroup.com/ [nuigroup.com] I recently built my own touch table from scratch - some IR lights point at a perspex surface, and an old projector back projects onto the perspex to provide a picture to interact with. The webcam has a small IR filter attached to the front, and this cuts out the regular lights. When my fingers touch the surface, they create hotspots that are tracked (known as Diffuse Illumination). You can also put

from tfa: it kinda works

[192939495969798999 (58312)]

If the video crashes at some point, a quick recap: it kinda works. You can see at certain points, the images get dropped and it looks like it doesn't totally track perfectly with where the fingers vs. screen are. However, it is an awesome technology and idea... maybe with a couple of remotes you could triangulate more precisely and get that true 'minority report' feel... just what i need for my tri-monitor setup :)

Re:

[toppavak (943659)]

It seemed to me like the app was just lagging a bit, its possible that the demo was just put together inefficiently on a slow computer- the other demonstrations by the original hacker were very impressive and smooth. I have no doubt that this could lead to an extremely well-polished and affordable interface in the not-so-distant future. You know what would be really amazing- if they integrated the 3-d "face" tracking hack into the system and built a 3-d/2-d hybrid no-touch interface. Moving around with the

Re:

[jma05 (897351)]

> its possible that the demo was just put together inefficiently on a slow computer

Highly doubtful. The objects move just fine after he has successfully targeted them. I think it has more to do with human performance. The issue is simply with the low motor resolution of the arm muscles compared to the more dexterous fingers. It can also be more fatiguing to hold them up in air since larger muscle groups are needed. MS Surface and Minority Report UI use large muscle groups to roughly locate the object but

Re:

[toppavak (943659)]

Highly doubtful. The objects move just fine after he has successfully targeted them. I think it has more to do with human performance.

Taking another look at the video, and the better quality one on their website, I'll have to agree with you on that.

They don't need efficient interfaces, just fun and cool ones... kind of like the Wiimote.

Thats exactly the point I was trying to make with that. With a no-touch multi-touch interface with position tracking using inexpensive hardware (a $40 wiimote, mostly open-source software, a few IR-LEDs that cost pennies and a pair of $5 gloves) you could create a really cool interface for a game. Even if Nintendo or a 3rd party developer doesn't come out with a game for the Wii that takes a

Pretty flash.

[Optio (1225470)]

I'd be very interested to see this kind of thing integrated with Bumptop [slashdot.org]. If Bumptop itself was then modified to have a little PC on it, well. Things start to get a bit recursive. Like standing with a mirror in front of you and another behind.

A desktop on a desktop on a desktop....

Wiimote

[MasaMuneCyrus (779918)]

Why not just use the Wiimote itself as a remote control? It seems more fun and practical.

Re:

[Bongo Bill (853669)]

Variety. Waving around an IR-sensitive, motion-sensitive stick is useful for some purposes that other interfaces can't handle, but the same is true of virtually any input device. This has its own niche to fill.

Touch and Feel User Interface

[Doc Ruby (173196)]

We're used to moving an actual thing around to do stuff. The physical reaction into our fingers is very important. The mouse gives a minimum, but the trackpad gives more. Touchless manual gestures don't keep the hands locked in a feedback loop with the virtual object, so they'll be clumsy.

What I'm waiting for is a thin memory plastic layer over a touchscreen, that can raise bumps and edges defining onscreen GUIs. Vibrating gloves could be good for simulating textures, but there's no tech for simulating tensile or inertial force in virtual objects. Maybe some kind of eccentric gyroscope, but I've never seen one.

Re:

[Moonpie Madness (764217)]

This is generally true, but we've seen lack of tactile replaced fairly effectively by clicking noises inthe ipod, and I see no reason why we can't do something similar here.

Or you could have a string that retracts when you manipulate something, simulating a force against the finger. That would be pretty sweet. Though I kinda find the whole project annoying. I like a cursor, mouse, and keyboard. I don't know how much time I want to sift through my pics, and keyboard shortcuts are way more efficient than

Re:

[Doc Ruby (173196)]

The clicking sound works because it's coming from something physical in your hand. The sound waves are close enough to actual mechanical vibration of your fingertips that the illusion in your ear (and in fact a tiny vibration of the fingertips) can work. But if there's nothing in your hand, the illusion from the learned association will fail, or be so tenuous as to offer a negative learning experience. Though I'd like to see tests of a demo with really high precision stereo audio apparently originating the

Re:

[Doc Ruby (173196)]

Well, memory plastic fibers do that without as much heat inefficiency. And though "addressably rigid" fibers are appropriate to simulating shapes of objects held in the gloves, they can't simulate position or other inertial values (eg. the motion of a ball through the air resisting the momentum of a catching hand). I wonder whether three elliptical gyros could cycle in proportions that both net to a 3D vector and sum to an inertial mass effect. If lots of those devices were nanoscale at very high speed, the

Re:

[Doc Ruby (173196)]

Maybe a ball with a surface covered in the textural memory plastic I mentioned, but elastic, with coiled shape fibers inside that can create textured shapes and sizes on demand, both convex and concave.

I really think there's some kind of gyroscopic way to simulate at least the inertial mass, and maybe somehow a 3D motion vector. The other option is some really dynamic electromagnet, but I never like the idea of magnetic fields sprayed around a room, especially one that can have magnetic media or CRTs.

There'

Re:

[Doc Ruby (173196)]

Thanks. Posting in the thread destroys the mod. Besides, this is all purely virtual, anyway.

Re:

[SuiteSisterMary (123932)]

Heat's not much of a problem, once you get the Clan heatsinks. Just don't get your myomer formulas from anybody named Justin.

Updated video posted

[cynergylabs (1225520)]

An updated version of this video has been posted to the Cynergy Labs Site. http://labs.cynergysystems.com/ [cynergysystems.com]

More info, better video

[pixelcowboy (617065)]

A better explanation of how he built this can be found here [cynergysystems.com] and a better video with a cool example of navigating a 3D object can be found at the Cynergy Labs [cynergysystems.com] site.

Where is the software

[brent_linux (460882)]

Ok I get it. You can make multitouch interfaces. I have made multitouch interfaces myself using a couple different methods.

They aren't worth a damn though unless you have something to use them with. Where is the multitouch picture organizing software that I can display on my coffee table and let me family sort through the pictures. Where is the multitouch D&D program that will let me and my friends move our characters through a dungeon with miniatures? Where is the multitouch coloring book that I can put a bunch of kids on? Multitouch math races? Multitouch Chemical Compound manipulation?

We need software. We have ways to interact now. We need things to interact with.

Two words

[Reality Master 101 (179095)]

Two words: tired arms.

Unfortunately, these sort of interfaces suffer from the same problems that doomed touch screen and light pens 20 years ago ("They can just touch the screen! How easy is that??") Users liked them at first, but holding your arm up is tiring. Try reaching out to your monitor and trace your Slashdot window for five minutes and see how long you last. It's *hard*.

There's a reason people in the Old Days wrote on flat tables, and didn't write on easels. That's also why artists who do use easels typically do "stroke and rest" (and why cartoonists use a flatter table)

A touch table is far superior for this sort of thing for that reason.

Re:

[slapout (93640)]

That's the same reason that voice activated computers won't fully take over: people get tired of talking.

Rumble

[spyrochaete (707033)]

I hope the gloves incorporate rumble. That's my favourite aspect of the Wii OS. Feeling that little bump when you scroll the cursor over a button is so tactile and tangible. It reinforces that you should immediately pay attention because you're about to execute a command.

Only problem

[floki (48060)]

As is well known in HCI research using your hands like this for some time becomes very tiring. But for showing off it's an impressive application :-)

It isn't Microsoft Surface!

[notanatheist (581086)]

Please stop referring to any multi-touch device as being like the "Microsoft Surface". MS did not come up with the idea of a multi-touch display. They steal and buy 99+ % of their technologies. Let's get PC and stop giving credit where it isn't due. It is a "multi-touch" surface. Not an MS Surface. /rant

Wiimote Turns Father into Touchless Person

[waveformwafflehouse (1221950)]

Perhaps those eight days would have been better spent caring for the screaming child in the background.

Not.

[no_opinion (148098)]

I got to play with the Microsoft touch at CES. This is nothing like it. The MS table uses a camera underneath the screen, so it can do things like recognize physical objects. Imagine thumbing through artwork on the table, then putting your wireless MP3 player down on the table and dragging the artwork to it and having it wirelessly sync. Pretty cool, if you ask me. They demoed this at CES. Basically anything with a barcode can be recognized as a unique device. Without this type of physical object recognition, the Wii version is a poor substitute, besides the fact you can't actually use all 10 fingers (or 20, if there are two of you) at once.

Wiimote

[Dan East (318230)]

In case anyone wondered, the Wiimote can track up to 4 infrared sources. The Wiimote's on-board hardware does all the heavy lifting as far as processing the image and determining actual coordinates (and sizes) of infrared sources. If a project only requires tracking four objects then the Wiimote makes a fantastic piece of hardware for experimental and hobbyist use.

So in this demo, all the manipulation is done by tracking four coordinates grouped into two pairs.

Dan East

Nintendo get moving

[MrCopilot (871878)]

A piece of $4.99 hardware and Jonny Lee Chung's Hack [gizmodo.com] creates the #1 best selling Metroid of all time.

Who Cares??? 3D VR!!!!

[Seraphim_72 (622457)]

http://www.youtube.com/watch?v=Jd3-eiid-Uw [youtube.com] Good lord this is cool. And he is right - bring on the games!!!!

urgh

[Tom (822)]

Interesting, except for the constant microsoft advertisement. "I love flash, but we built this in silverlight (continue on to long rant about how great that is with no relation to the topic whatsoever)", or the "and since it's built in .net it can communicate with the Wiimote", err yes? What's that gotta do with .net? Then the "oh, look a picture of me at some microsoft meeting", and on and on. All that really got on my nerves about 3/4 through the video.

Re:If only we could control Slashdot with a Wiimot

[by Anonymous Coward]

If you saw it three weeks ago, then why didn't you post it? Oh, you were too lazy? Then STFU.

Re:

[garcia (6573)]

If you saw it three weeks ago, then why didn't you post it? Oh, you were too lazy? Then STFU.

Because I didn't think it was worth discussing on Slashdot? Perhaps because in the past when I've submitted what I thought was pretty cool shit it was rejected within minutes or rejected and then posted a few days later by someone else instead that had, what I felt to be, a lame writeup?

But most of all it's because I can't stand the fact that some of the writeups are nothing more than blog advertisements that link

Re:If only we could control Slashdot with a Wiimot

[Cheapy (809643)]

If you saw it a few weeks ago, why didn't you submit it to slashdot? Your post makes no indication that you did so.

Slashdot is like OSS. If there's something you don't like, try to fix it yourself. At least that's what I've been told dozens of times.

Re:

[AKAImBatman (238306)]

Uhhh... no. This article links to a previous article, which links to the article you linked to. Each of these articles shows a progressive evolution of the concept. The article you linked to used reflective tape to accomplish the IR tracking. The article linked to by this article shows how to use a light pen for greater accuracy. THIS article combines the two approaches using IR equipped gloves to create a highly accurate touch surface.

Not a dupe- sort of.

[toppavak (943659)]

The first article was about the original hack which used a test interface and IR-reflective tape. This article seems to be about a project to implement the no-touch hack in a more usable fashion in an actual application. Definitely a step forward and pretty cool.

Re:

[PrescriptionWarning (932687)]

Let me guess, you just lost on the gameshow "The Apprentice" and you feel sore about it.

PS. YOU'RE FIRED!

Re:

[techpawn (969834)]

And no thought magic for me please, hand is the highest interface I wanna go.

Must... Resist... Joke...

Re:

[Constantine XVI (880691)]

Get this ported to OSX then. OSX has the Newton's handwriting system under the name of Inkwell. You just need the proper hardware for it.

Re:

[randyest (589159)]

Well, Bluetooth doesn't require a table. Seems like it would be trivial to just hold or point to a Bluetooth-enable device like a portable hard drive or camera, then have the data transferred. What's missing again?

Re:

[randyest (589159)]

Why would the device need IR lights on it? If it's bluetooth, it can be detected, and the system can be set to make a gesture or menu selection to get data from it. If there are more than one bluetooth devices in range, you'd have to select which one. Just like the Surface, which requires you to apply barcode-like stickers to each device so it can be recognized/distinguished from other bluetooth-enabled devices in range.