Open-Source, Multitouch Display

Comments

• Correcting An Error fennecfanatic on 05/03/2008 at 5:30 AM

  Posts: 2 Avg Rating:

  Re. "...a simple webcam with an added infrared filter..." Actually, the necessary modification to the webcam is to remove its infrared filter and add a visible light filter. Almost all webcams come with an infrared light filter, which would prevent the infrared interface light from reaching the lens of the camera. Swapping the filter causes the camera to see only the infrared light reflected by objects on the touch surface, blocking out interference from projector reflection. Such visible light filters are typically made in the do-it-yourself community by cutting a square from an exposed area on a film negative. Rate this comment: (Reply)

  • Re: Correcting An Error feign on 05/05/2008 at 2:14 PM

    Posts: 1 Avg Rating:

    Actually there are a number of cameras which do not come with this infrared filter, the apple iSight being one of them. I built a multi-touch table very much like the one described here for an independent study credit a couple years ago, and I was actually surprised to find that all of the webcams we played around with were able to see near IR right out of the box. To turn them into purely IR cameras, all we had to do was tape a visible light filter over the front. In my experience, film negatives didn't work too hot, but if you want a DIY solution, you can disassemble a remote control and use the filter that covers the IR LED. For just a few bucks though you can buy a decent plastic one from photography stores and probably science supply surplus type places if they have a decent optics section. Rate this comment: (Reply)

• Prior Art dandb on 05/03/2008 at 8:54 PM

  Posts: 1 Avg Rating:

  In the late 1980s (1988 I think), I saw a multitouch display at Bell Labs built by Bob Boie.  It used a transparent capacitive array and had excellent speed and resolution, I recall.   I probably have more details in my PhD thesis (The Automatic Recognition of Gestures, Carnegie Mellon Computer Science, 1991) but I'm not currently near a copy. I got to talk to Bob Boie for a while.  The guy was amazing.   He seemed to turn out incredible inventions at a pretty regular pace.   No one at Bell Labs that I talked to seemed to know what to do with the multitouch pad.  I recall they had two ideas: (1) Shift Keys(!), and (2) step on the pad in a shoe store and get a detailed pressure map for your foot, so the salesman can sell you the expensive shoes. For the multitouch part of my thesis, I used the Sensor Frame, invented by Paul McAvinney, then of Carnegie Mellon, later of Sensor Frame Corp.   It could detect up to three fingers in the plane about a quarter inch above the screen. It used optical sensors in the corners of the frame, so it had trouble disambiguating certain finger configurations.  Anyway, I saw in a prototype hardware around 1985, I used the technology (just a single sensor)in the VideoHarp around 1986/87, and I used a working Sensor Frame in 1990/1991. I ended up using the Sensor Frame to make a multifinger drawing program, which was first mentioned in SIGGRAPH 1991, with lots more details in my thesis, and a video in the CHI '92 video proceedings.  The program had a two finger gesture similar to the iPhone: pinch to shrink, separate to enlarge, rotate to rotate. The author Ms. Greene is right when she says the multitouch technology has been around in labs for decades, though the NEC example she cites seems to have first been published in 2000 or so.   I hope the two examples from the 1980s I mention are of interest. Dean Rubine Rate this comment: (Reply)

• 	oless abb8404 on 05/05/2008 at 11:49 AM
  Posts: 1 Avg Rating:	you quiero construir unaa Hello My name is Alejandro Bonilla I write from Cali - Colombia I am interested in the cube that is what it takes to build I would like to build one for my home; about this issue please could provide information? thank you very much Rate this comment: (Reply)