Hi, all.
I want to thank all of you for the useful information posted on this site.
I’ve just started a multi touch project for my invention, and would very much need any support that I can get finishing this.
I have built a virtual projection screen and for that I need a virtual control. (open air)

For the virtual touch part I will attach a simulation photo. Basically I used 2 cameras for triangulation. And using them aligned with the IR wall separates the light to only that from the IR emitter. Than can be an array of IR LED or a IR laser that scannnes a plane.

I am having trouble with the software implementation. I have no experience in this field. But I caught on very quickly.

My current setup features a 10cm long IR grid with 16 IR LED and a webcam with IR filter in front of the grid.
I managed to get some results with simple contour track, vvvv patch posted on this forum, but after closing the application the image remains darkened.
Also I got a better result with reacTIVision. But I do not know how to implement this with any software to actually control something.

Your help is much appreciated.

nice video, nice idea, but i don’t get the technique yet that good, are the 2 optical imaging devices the camera’s? can you be more specific about how this works, looks like a nice idea

Wow, that is some awesome screen you produced!!

Not so sure about your interface idea, don’t see the grid thingy, but could be wrong.

Thank you for your reply.
Let me explain you more about this concept.

I have attached an image

1. Base
2. Two cameras that are used to capture the interaction between the IR grid (3.) and user’s hand
3. IR grid
4. Field of view of the camera

So, a beam of IR radiation is emitted from below the base, either from a line of IR LEDs or a scanning IR laser.
In front of the camera’s lens is placed piece of metal to narrow the horizontal field of view (FOV), and letting the camera’s FOV just vertical, to filter the outside IR radiation. In this conceptual setup i used two cameras to triangulate the point(s) of interaction. Coping for the third dimension given by placing the camera in front or back of the IR grid. By placing the cameras on the side facing each other helps to achieve a very narrow setup.

I hope that this explanation will make you better understand my idea. I am sorry for my lack of knowledge of English technical terms.

**Please tell me how to make the google video ( enbeded ) viewable directly from the post on this forum.

Have you tested this out at all?  I would be worried that the IR light would be blocked/reflected by the water vapour/smoke making it hard to register a finger.

Yes, i have tested. I have a current setup of 10 cm length.
But I am having trouble with the software.
For that I need your help.

thanks

Hey Holopix, interesting project you have here, you might want to look into using some IR sensors (see attached) compared to cameras, however I do not know of a grid like this allowing for a multi touch environment. As far as the software, what would you need help with? What kind of cameras are you using to detect, and what software?

Is your setup suited for multitouch? I was thinking about connecting a bunch off these (or simulair) things, but you would need loads to get a decent resolution, so not much off an option. (and looks prety $$) and it isn’t MULTI-touch.

A multitouch gid floating in air.....  hmmzz.. even more intriging than the FTIR things we are trying to build

As for software, sry, I only know vvvv.

But I hope to see your blog on the particle screen soon!!

I am sorry for this late reply.
I have being working on the projection smoothening it’s surface
Thank you all for posting. I am sticking to mine because of the similarity to FTIR.
Can anybody put me up to speed with the this? Which software is the used successfully for multitouch?

Here are some videos:

http://www.youtube.com/watch?v=O2_MUxNDC2g
http://www.youtube.com/watch?v=IJh8zYZsiAc

thanks

well all kinds of software are used with succes, VVVV, processing, flash, java, c++

currently there are some improvements being made in touchlib (c++) the open source multitouch library that whitenoise wrote

maybe you can find some answers to your questions in the software section

can you give a bit more details on how this actually works, you beam an image on the IR grid?