Han suggests using Rosco grey projection screen as a compliant surface. I ordered some of this, and in my experience it makes a terrible surface, requiring excessive force to trigger off the FTIR effect.
In fact, if you look at these flickr photos of Han’s latest prototype he hasn’t used Rosco grey at all. What it looks like to me is he has some kind of rubber (looks reasonably thick - much thicker than a projection screen material)) which couples with the acrylic under pressure (setting off the ftir effect) and this rubber also seems to be translucent enough to act as a projection surface. On top of this rubber he has overlaid a clear sheet of plastic to provide a smooth surface for your fingers to touch.
I have experimented with coating the back of projection surfaces in rubber, and have found silicone sealant works very well. But it tends to stay sticking to the acrylic once you release your finger (causing the light to still be scattered), to solve this I tried adding a fabric/rough texture to the silicone when I applied it. This seems to give the silicone the property to stick to the acrylic under pressure but to push back to its original form when that pressure is released - this works very well. The only problem with this method is that it’s difficult to apply to a large area evenly, and the silicone tends to peel off the projection surface if damaged at all. So an already manufactured solution would be much more desirable.
I did think about using silicone sheeting (usually used for gasket cutting), as this type of silicone appears to be a bit stiffer than the silicone sealant, and as a result might not permanently stick to the acrylic (if it sticks at all!). But I don’t really have the budget to keep randomly experiment with materials. It is also possible to get silicone gasket sheeting with a textured finish (although I only found 1 example of this in red oxide colour).
I raided my household for everything plastic and rubber (packaging, pen grips, car matts, shower curtains etc) and found only silicone seemed to produce the FTIR effect under normal pressing/touch force. But there might still be materials out there that work, its just a matter of finding them (eg I haven’t tried latex).
If anyone has spoken to Jeff Han it might be worth asking him what he used (he didn’t reply to my email, but if enough of us ask he might tell?).
So anyway, post your compliant surface experiments!
Oh and if you’re wondering “why bother with a compliant surface?” I think it offers these advantages:
1) Protects the expensive acrylic from scratches
2) Blocks a lot of light pollution.
3) Provides consistent results (the effectiveness of the bare acrylic touch seems to be down to how sweaty/greasy yours hands are).
4) Zero visual disparity between the touch surface and the projection surface
5) Pressure sensitive
6) Seems to react better for dragging movements (at least in my experience)
7) Brighter blobs to track, as there is no longer a diffuser between the IR blob light and the camera.
8) World Peace
I ordered the exact same Rosco screen (as written in Han’s paper).
At first I wanted to try and use it as a diffuser below the acrylic, but you definitely want a compliant surface overlay on top of the acrylic.
I hope to receive my Rosco screen pretty soon, so I can fool around with it. I’ll let you about my experiences.
So what kind of material are you going to use to project the image on from the projector.
I mean, like I see it now, your setup is like this, with the first one being on top:
I’m currently using a white shower curtain as the projection material (overlay material in your diagram), very cheap, but they have creases which are difficult to remove. I might use the rosco screen for the final version. The rosco screen is probably too thick to be used as a diffuser below the acrylic.
But like i say Han has seemed to have found a material that can be used as the projection screen and sets off the FTIR effect.
Have you tried projecting an image on that shower curtain with a projector, or have you just been playing around without a projector?
If you’ve been using a projector, was the quality of the image shown on your curtain good or not?
I got the post of mine of the touchlib mailinglist, maybe people out here have the same question, eventhough the answer maybe is allready given:
I already asked around in #ftir, but I could use some more opinions.
I ordered a piece of rear projection screen of Rosco which should arrive
within 2 weeks now.
But I was wondering, should I use it below the acrylic and use another
compliant surface overlay on top of the acrylic or should I just use that
screen as a compliant surface overlay and don’t use an extra diffuser below
the acrylic. Normally I’d say, just the acrylic and on top the Rosco screen,
but the screen can get very dirty I think or it can get cut, stuff like
that.
What methods did you use, maybe, the Rosco screen as a diffuser below the
acrylic and on top of the acrylic a different compliant surface overlay?
Because the material of acrylic is pretty raw, what I mean to say is, your
fingers don’t easily slide over it, the blob gets less visible. I saw people
using some kind of sylicon or gelly substance to make the surface
slippy/smooth, but in the end you dont want to let the user end of with some
kind of gelly substance on their fingers.
Don’t buy an expansive rosco screen as a compliant surface overlay!
today i received my rosco screen.
i cut it into pieces and started testing....
and the same thing happened to me as to tj_joobs, as soon is I put the Rosco screen on top of the acrylic the FTIR was very very minimal when i pushed as hard as i can
so i wouldnt suggest the solution which was mentioned in Jeff Han’s papers to anybody
I havent tried projecting on it, maybe it’s good for projection but it sure isn’t for giving a good FTIR in combination with projection
soon i’m gonna buy some silicone spray to put between the acrylic and the rosco, because tj_joobs had experience with that and he told me that this would improve the performance of the FTIR
I think i’m also going to look around for latex and sylicone surfaces to put between the acrylic and the rosco because of durability, because in the end you dont want some sticky stuff on your screen
the best thing would be to find a material that works as both a as projection surface as well as giving a good ftir effect
let us know if you find a suitable solution
ps: touching the screen with a kind of surface compliant overlay like this is better though
I did think about using silicone sheeting (usually used for gasket cutting), as this type of silicone appears to be a bit stiffer than the silicone sealant, and as a result might not permanently stick to the acrylic (if it sticks at all!). But I don’t really have the budget to keep randomly experiment with materials. It is also possible to get silicone gasket sheeting with a textured finish (although I only found 1 example of this in red oxide colour).
So anyway, post your compliant surface experiments!
Hi
I just bought a silicon sheet, silicon plate and silicon spray. All of them seems not to work better than the raw acrylic.
The silicon sheet(basicly paper with one side coated with silicon) is pretty useless in my case because it’s just silicon on one side. The silicon plate is very thin, soft, flexible, very smooth, has a milky tone to it and is (very subtle)textured on one side.
The spray stays smeary on the acryl, so it’s useless too :(
I also bought a thin piece of translucent latex, which is nicer to touch and seems to make a great diffuser and projection surface. I haven’t tested it to the detail, but I would claim the light scatters with a tiny bit less pressure than it would with the bare acrylic.
By far the best thing that works is plain watter or even better: thin layer of water under some translucently plastic foil. But this doesn’t make a good permanent solution :(
I wonder if a thin layer of oil under the silicon or latex would help(water didn’t).
@Joobs: could you post a picture of your setup? I also tried the silcon sealant. I have put directly a thick and a thin layer on the acrylic. The thick one is so uneaven that it reflects the IR light without pressure in some areas. The thin one is cool and works OK though is also too noisy and breaks really fast.
I wonder how your fabric texture looks like.
I’m a bit confused by your post. could you possibly take some screenshots of what the camera sees from each of those surfaces? Most surfaces layed ontop of the acrylic dont set off the FTIR at all.
silicone paper - you mention is only has silicone on 1 side? You only need the silicone on 1 side? the idea of the silicone is to simulate soft flesh, so this goes between the acrylic and the projection surface (because most projection surfaces dont set off the FTIR on their own).
Silicone plate - This sounds ideal to me? how well does it work?
Latex - This sounds like it is working well too?
It doesnt matter how rubbery they are to touch, because you can put a clear plastic sheet over the top.
I wouldnt reccomend using a layer of water and then a surface ontop, as this tends to permanently set off the FTIR.
You cant actually see anything from the picture of my compliant surface, its a shower curtain with a very fine layer of silicone sealant on one side. You all you see is a shower curtain with one side a bit more sparkely where it might catch the light.
I’ve made a diagram though to make it a bit more clear
today i had a go testing the rosco screen for projection, which works very well
but when you put the rosco screen on top of the acrylic there is hardly till no ftir effect
when i place the rosco below the acrylic the blobs are very vague, i even think that software cant help that
we also used a piece of tracing paper on top of the acrylic, which doesnt give ftir effect at al
and when we put it below the acrylic, the blob were very dificult to detect and the projection onl the tracing paper was less clear then the rosco
so if you have some budget, i suggest a piece of rosco instead of tracing paper (a3 format = 25 cents) for the projection, but still the FTIR is not good enough
(in the end of course you want 1 layer material to put on top of the acrylic which is good for both ftir effect and projection at the same time)
so i think we will try with some silicone material between the acrylic and the rosco screen next time
I’ve bought IKEA window foil (to make your windows appear matte) and have been using this for a while now.
When the foil is used a lot and not sticky anymore I used it as a top layer, the foil scatters a lot of light but still allowing the camera the see the fingers (and the lit hand/arms, but when you alter the threshold it should work better).
Using the foil underneath (that’s how I use it now) gives the best results for me. Instead of acrylic I use plexiglass(cheaper) as the FTIR panel. And heck I don’t even use IR lights (I use christmas lighting...so I imagine it would work much better with real leds)
The raw plexiglass surface however is still somewhat hard to glide on with your fingers, but since it is a bartable-surface it gets wet and then it goes a lot better.
It would be really good if you guys could create some kind of comparison videos/images, showing how youre compliant surfaces compare to the bare acrylic.
For those who havent seen it this is my video, I would say that my silicone sealant experiments shows slightly less sensitivity to direct touch but increased sensitivity to dragging. Also allows for pressure sensitivity.
Yesterday I handed in my prototype for evaluation (I made this multi-touch project for a course I take at TU Delft), so I cannot make any new videos. If the attachment thing is up I can post some pictures or video’s I made earlier.
You’re right, plexiglass is acrylic, so is perspex and altuglass all brandnames. I am using the an extruded polyester plate - PETG also known as Vivak, Vectan or Hostaglas. We use these at our faculty for vacuum molding. There is a significant price difference of 30 euro’s for a square meter (€105 for PMMA and €75 for PETG).
Also I would like add that I used the nonsticky part of the IKEA foil at the underside of the panel. If you stick the foil to the plate it obviously renders the FTIR useless.
(I use christmas lighting...so I imagine it would work much better with real leds)
