The benefit of the DIY versions of these types of projects is that the technology and plans are available on the internet for anyone to access and you can build one for a fraction of the cost of the professional products. Of course if you want better performance, you need to shell out more money. The table I built for MAGFest was the lowest budget I could manage(around $800), but still performed fairly well even though it had some bugs such as not recognizing small touch surface area at times and misinterpreting the exact orientation of the symbols on the cards. Those problems could all be solved with better equipment.
A video I took of people using my table
The technical term for my table set up is Rear Diffused Illumination. I learned a lot about what I know from videos on youtube, the NUI Group forums, and bought some parts from the Peau Productions store(same guy from youtube). I bought low cost infrared light emitters that shine up through the table surface which is comprised of a rear projection material stretched over a home depot bought acrylic. Anything that touches the surface reflects that light back down into a Playstation Eye camera that has been modified to see only infrared light. The PS Eye sends images to a program called Reactivision which recognizes symbols on the back of cards and finger tip sized blobs, registering their number and orientation and sending that information to an Adobe Air program written in Actionscript 3 using the TUIO AS3 library. The Air program is then projected onto the surface rear projection material. The program I used at the show was running the standard Flash engine, which I have learned is rather clunky and since learned the Starling game engine for flash which purports to make use of full 3D hardware functionality allowing for greater performance on a variety of devices.
In Development video
Before this video was made I had a phone call with Cortina Productions and scheduled an interview. I wanted to build something to bring to the meeting but only had 5 days. The interview then got delayed another 3 days so I had a total of 8 to get it running the above video shows the end result. Spoiler alert, I got the job. The video below shows the inside of box and early progress.
This is a functioning concept for an interactive kiosk. This project spans not only the front end user experience but also the content management side of how to best get varied content from many sources. The video goes into detail at length, but the reason for this project was simply that I enjoy making interactives and I like the idea of providing a platform for market vendors to attract attention. At the same time it would give festival goers something entertaining and informative to play with.