First of all, lot of thanks to NUI Group, Fairlane, MachineGun and all of you.
I have in mind to build differents setups, and with that, to develop some appz’s: so simple, so complex. The first proyect we just did it’s a Rear DI setup (we are from Technical University Federico Santa María, Valparaíso, Chile: http://www.usm.cl, and particullary from http://www.idpusm.cl). We will build an FTIR very soon, and an LCD too. With my partner we have good ideas about appz’s, interfaces and interactions, we’ll post about this when we have something to show (simulations or appz’s working on multitouch).
About our first table, Rear DI:
.- Projector: BenQ522ST
.- Mirrors: two normal mirrors
.- Camera: PS3 with 850 [nm] filter (from http://www.peauproductions.com/)
.- Diffuser: Shower Cortain Plastic (over Acrylic 8 [mm], 120x79 [cm]) Acrylic because we have planned to use the same for the FTIR setup (soon)
.- IR light: 4 Lamps with 42 IR LEDs each one (850 [nm] of course)
.- two fans
Some info about our building process:
.- We choose to make the box with a trapezium shape, I still without know if was a good decision, because we have to use two mirrors; we are using 2 normal mirrors, so we have some of ghost image.
.- We can’t get a better diffuser yet, like Rosco Grey or 7D006, so we try tracing paper, a piece of frosted glass and ‘shower cortain plastic’. Some test with frosted glass v/s shower cortain plastic in this thread (started by dannyzs): http://nuigroup.com/forums/viewthread/6527/.
.- The surface it’s 120x79, but our useful projected area it’s 90x66 [cm] aprox. From that area, we just can use like a 60% :(, I’m almost sure that’s because we aren’t using the camera with a wide angle lens, please confirm this if you have knowledge about this(Update: see post below, with vids). Today we try to put the plastic over the acrylic very carefully, to get surface tension (we use double contact tape); after that, we could get a respectable blob detection.
.- The quick way to do IR Illumination it’s just to buy lamps with IR leds, but Mr.T http://nuigroup.com/forums/viewthread/4401/ and zviangi’s Table http://nuigroup.com/forums/viewthread/6508/P0/ (video: http://www.youtube.com/watch?v=JZM5p0D7ULk) use LED’s arrays, and with that you can get even IR illumination definitely (thanks zviangi’s for the feedback).
Things to do (in this proyect):
.- To find where we can get Front Surface Mirrors in Chile, to get a better image without ghosts.
.- To find where we can buy Rosco Grey in Chile, or a supplier who can shipping to worldwide, or another good diffuser not so expensive.
.- Try a wide angle lens, find where to buy one [Update: see another post below, with vids)
.- Maybe to build led’s arrays?
(the final picture in the video it’s my university )
and some pictures below
We have a long way to keep investigating, developing MT devices and appz’s, I hope this thread be helpful to the newbies. Any observation or question will be very welcome!
I have approx the same size of the screen and I am using wide anlge lense, before, with 43 degree lence, I could use only a part of the table, like you.
one hint (maybe will help wit IR light): try to put some white paper on the walls, inside, or some reflective paper, or silver scotch, it might help to diffuse more IR light, espesially, when you have your illuminators aimed to the walls
May I asked from your pictures, (1) I saw one mirror reflect image from the projector, what about the second mirror, what there for?
(2) I also see another light source beside the IR emittors, what there used for?
have you think use metelic coating like emergency car kits Reflective Blanket for IR reflective material and cheap, it reflects IR better than white light?
May I asked from your pictures, (1) I saw one mirror reflect image from the projector, what about the second mirror, what there for?
(2) I also see another light source beside the IR emittors, what there used for?
have you think use metelic coating like emergency car kits Reflective Blanket for IR reflective material and cheap, it reflects IR better than white light?
@uncledj: that light is just a reflection of light bulb from the room, there is a clear acrylic over the top and that reflected it, verdad?
mirrors: as I got it, first - small mirror gets the image first, reflects to the bigger mirror, which then reflects the image to the screen. interesting placement of mirrors, I havent seen this way setup of mirrors.
@ Peau: Thanks for the good energy (and for the PS3 of course )
@ ajlovegrove: exactly, trapezium it gives you a relatively small box with a nice shape, and you can be close to the table without a bag’s pain… but the kind of projector (distances, short throw and that stuff) puts you restrictions = number of mirrors; but for now we’ll going to keep that shape more about ‘trapezium shape’ here http://nuigroup.com/forums/viewthread/6328/. more cameras working simultaneously means less box high too.
@ zviangi: thanks, I’ll take your advice about lens and some reflective/diffuser material on the walls, definitely. Your led’s arrays rocks.
@ swordrift: fans are not connected to the computer; I’m very bad to deal with wire things, so my partner do the job, see picture attached please.
@ uncledj and zviangi: (1) like zviangi says, it’s just a bulb reflected; and (2) to get that trapezium box size, we had to calculate the position of mirrors; we test a lot, but it was great help to use this appz (http://robertpriewasser.com/stuff_projects/programming/rearproject/shockwave/index.html), it help us to get aproximates distances and angles. Some more pictures attached.
Hi Fernando , so did you find a solution for this problem ??
Hi! I could say yes, Soon I’ll upload a video. This is my situation:; I didn’t make the box with a perfect planification, so I can move IR lamps and camera and everything, in a limited space. But yesterday I try another position for the 4 IR lamps, and I get relatively good blobs in all the projected area (compared with my last videos). Now the problem it’s that blobs become weack at the edges, but less than before; I mean, with the new position of IR lamps I’m getting the best performance for a MT with ‘shower cortain plastic’ like diffuser.
In fact, just now I wrote this in the chat: ”Hi; Anyone knows how can I get ICE Holocontra? I wrote to link to mail and I’ve no answer yet (I wrote since 4 days ago). There is another way to get ICE Holocontra? (with shipping to Chile) Dou you recommend ICE Holocontra or 7D006? thanks nuigroup” UPDATE: just now I PM to ICEAV nuigroup member
I promise to upload some videos with my last setup
Fernando, great design!! Love the trapezium shape. I was thinking on building my setup on that way, but im currently prototapeing a regular table couse i dont want to use mirrors. How about the keystone effect with that setup??
Y ahroa en español! uff, que alivio! buenisimo el diseño! te felicito!. Quería armar mi DI de ese modo pero tengo problemas con el espejo (por ahora probé con uno solo) y la imagen se me deforma (keyston). Tambien tengo un benq ST, pero en mi caso el mp512st. Con esa configuracion te queda bien o la tenes que corregir desde el proyector?
did you try with the white papers on the walls? the problem with the blob at the edges, as I think is a problem of the reflection angle from your fingers: imagine: your fingers in the center of the table, the IR ray is bouncing back from the finger with the angle more or less visible to the camera, because camera is in the center and finger also facing the IR rays from the center, as the illumination in the center is more intensive. For edges, you have rays going to the finger from the center and reflecting toward the wall, if you had a good illumination from the wall, toward the center, then the most reflections would go toward the center and camera… entiendes lo que digo? I have the same problem with the edges, blobs go weeker at the egdes, because effective reflection goes toward the walls.
Si usar la fisica, estare claro, porque rayos ban a otra direccion, porque los reflejan con mizmo angulo de caducar a dedos.
Fernando, great design!! Love the trapezium shape. I was thinking on building my setup on that way, but im currently prototapeing a regular table couse i dont want to use mirrors. How about the keystone effect with that setup??
Hi Diego; I will write in english to practice
I am glad to know that, at least the shape has been appreciated, thats good.
About mirrors: I am not having issues with keystone, maybe just a very little that can be fixed easily with MP522ST keystoning functions; you have to calculate distances and mirrors, here in nuigroup there is post about diferent ways to do it, like to use this: http://robertpriewasser.com/stuff_projects/programming/rearproject/shockwave/index.html, or you can search “mirrors” or “projector”, that must be give results, threads dedicated to this. Actually there is a paper technique to do good calculations. After that theory, Its better to test that angles and distances with mirros, and/or paper and projector before to build the box.
saluos che XD
zviangi - 12 August 2009 03:44 PM
Hola Fernando,
did you try with the white papers on the walls? the problem with the blob at the edges, as I think is a problem of the reflection angle from your fingers: imagine: your fingers in the center of the table, the IR ray is bouncing back from the finger with the angle more or less visible to the camera, because camera is in the center and finger also facing the IR rays from the center, as the illumination in the center is more intensive. For edges, you have rays going to the finger from the center and reflecting toward the wall, if you had a good illumination from the wall, toward the center, then the most reflections would go toward the center and camera… entiendes lo que digo? I have the same problem with the edges, blobs go weeker at the egdes, because effective reflection goes toward the walls.
Si usar la fisica, estare claro, porque rayos ban a otra direccion, porque los reflejan con mizmo angulo de caducar a dedos.
ha? how is my spanish?
Hi zviangi; your spanish is not so bad jajaj, I look your blog in georgian and I saw characters that I’ve never seen before! XD.
About weack blob detection at edges: I did’nt understand completely that stuff about angles and camera and fingers, but some of that I could get anyway, and it makes sense; but there is another stuff about lenses: maybe I’m using a lens with focal distance that makes to get weack blob detection at the edges. My partner bought a 1.8 mm m12 lens, and maybe I need another one, like 2.1 mm m12 lens; In that sense, Peau did a lens calculator http://www.peauproductions.com/blog/2009/07/23/m12-lens-and-distance-calculator-formulas/ I did use this but I can’t understand why I get two different values for lens (attached picture of excel).
But my great problem I think it’s the bad diffuser; I’ll buy another one, ICE Holocontra or 7D006. I’ll post videos soon to show the new position of IR lamps and screen shots of what sees the camera in CCV, see ya!
)
more about ‘trapezium shape’ here 
