Oh, Hi! I didn't see you come in...

So yeah, about my MT table...

As you can guess by my long absence on this blog, I dropped that project a while ago, mostly a question of time and motivation. But I did not log my final progress so here it is :

PS3 Eye

Today, I bought a PS3-Eye (for 40€)
First, I was nice with it. I found a driver here to get it to work on my PC (it has a USB plug). And it worked well.
But then, I ripped it open and cut its inside with a knife.

The PS3-Eye can be used as the camera for a multitouch table (it's actually the favorite of the community). However, it has a IR-filter that you must remove first. It's actually quite easy to do, just do what the guy in these two videos tells you.

Why did I choose a PS3-Eye? Well you're supposed to get a VGA (640x480px) at 60fps, and a good framerate is important for responsiveness (I haven't managed to get 60fps, but I think it needs some configuration ; I'll look into it). Plus everyone seems to like it, and it's only 40€

I tried it with CCV and it worked fine, except that I don't have a bandpass filter yet (so instead of picking up only IR light, it can see IR and normal light, and that's not what we want).

A few more things about my prototype next post.

Still alive

Hey!
Just saying hello after that long absence...
I haven't have much time to work on ECHE, but now that I'm back, I should finish the prototype this week-end.

By the way, I just saw that video of a table, which is supposed to be cool because you don't have to actually touch the table...




It's funny how what some would call bad calibration can be advertised as a feature...

src: Engadget

External frame

I have managed to get something that looks like a frame for my prototype. It's a 1/5 model, but it's not to scale. The edges will keep the same size, so the ratio screen size / total size will improve greatly.
2 things:

• it's heavy! Not that heavy (it's made of aluminium), but I'm pretty sure it's heavier than an iPad.
• it doesn't look good. I will probably have to paint it

Update: No Update

Hey,

Not much happened this week-end, but

• I've designed the circuitry and chosen the components for my miniature screen. Cost: 15€. I may have taken more LED than needed (around 1 per cm). I'll give the details for the final screen, but sneak peak: I used this site to figure out which resistor to use.
• I received my mini plexiglas sheet: (the blue vibe is due to the protection)

• I found (actually borrowed) a useful tool to cut at 45° (to make the frame)

So it's metal-sawing time again, yea-ah!

I expect to finish the prototype this coming week, and move to the camera-testing phase.

In the meanwhile, I may write a small piece on multitouch and... phones

[insert signing out catchphrase]

ECHE: specs

My Multitouch table, that for some reason I call ECHE, is for now mostly in my mind, but here are the specifications I am aiming at:

• FTIR technology (see my previous post if you don't know what it is)
• screen size: 140cm x 90cm, which is a 66" diagonal (only time I will ever use imperial units, but they are the standard for screen diagonals). This is counting the frame, so the actual image will be a few centimeters smaller
• height: it will look like a drawing table, and the angle and the height will be adjustable, so you can use it standing or sitting in a chair
• acrylic: Plexiglas XT, 8mm thick with polished edges, sold by plexiglas-shop.com
• projector: I don't know, I liked the Acer S1200 because it's a short-throw, but I think a 1080p would be better. However I haven't find any short-throw 1080p projector, so I am still investigating the subject, I may use an HD projector and a mirror, or add customized lenses...
• camera: I was thinking of the PS3 Eye, because it's the most used cam in the community, but it's mainly because it is cheap. Since I'm gonna have to pay for a good video projector, I could put the extra buck for a better camera, I don't know yet
• projection surface: I have no idea for now
• Compliant surface: not sure yet, but a teflon coating sounds good.

The estimated cost is (keep in mind I'm in France):

• the acrylic sheet: 120€
• the projector: around 800€
• the frame: max 200€
• the electronics: max 100€
• plus every other things I can't price yet, and the hidden costs I will discover

for a total between 1,500€ and 2,000€, which is a big price, but it will be a big table. And I already have a buyer, so that's cool.

For now I'm working on a miniature prototype, to see if I can get the frame and the electronics right, and it's mostly about sawing and piercing aluminium, so I haven't any interesting photos yet. But stay tuned...

Now I'm gonna cross-post this on the NUI-Group forums, to say hello.

See-you!

Multitouch DIY for dummies

I've just realized that I am going to use the acronym FTIR a lot in this blog, and I should at least define it once.

The Multitouch DIY Scene deals mostly with techniques using Infra-Red (IR) cameras. Why? Because they are cheap and easy to find, and those techniques allow screens almost as big as you want. When you think about it, a digital camera is a lot of censors all wired up with the correct interface to easily connect them to your computer. The use of IR instead of standard camera is for the image of the screen not to mess with the image picked by the camera.

There are a few different techniques using IR, and I urge you to go there if you want to know more. Basically they all come down to this: there is a IR source somewhere, and when you touch the screen the IR light is reflected on your fingers and is picked up by the camera. Where you put the source is what differentiate the 4 main techniques.

In FTIR (Frustrated Total Internal Reflection), the IR light is beamed inside a plexiglass sheet (which actually is the screen). Then, when you touch the screen, some weird phenomenon creates a reflection that can be seen by the camera.

That is the technique I want to use for my table. Why? Because it doesn't use lasers, produce enough light to not need to be completely enclosed, and doesn't need a specific and expensive kind of acrylic sheet. Of course, it also have drawbacks, for example it cannot be used to detect fiducials. But I don't intend to use fiducials, so it's ok.

And now, you will know what I mean when I say that I want to build a FTIR table.