Multi-Touch Interface Research

Testing the surface. We accomplished to build a surface that allows us to register even very subtle touches. It has Rosco Grey on top, which is both an excellent projection screen and also very nice to touch. Beneath it we use silicone to trigger the FTIR effect.

Photo by gregorhofbauer on 18th March, 2008. 

Yesterday we got a brand new short range projector delivered to build it into our multi-touch table. That baby literally was just released to the market the day we got it, and we are really happy with the results. Andreas also did the first alignment tests and while it will be quite some more time passing until we got the perfect camera and software settings, the actual multi-touch-detection works pretty well. We love sending the green dots all over the screen, tied to our fingertips.

Gestern kam endlich unser brandneuer Kurzdistanz-Beamer ins Büro und wurde sogleich in unseren Multitouch Tisch verbaut. Dieses feine Gerät ist tatsächlich erst seit gestern im Handel erhältlich und wir sind recht glücklich über die Leistung bis jetzt. Andreas hat auch die ersten Ausrichtungstests durchgeführt und obwohl es sicher noch einige Zeit dauern wird, bis wir die perfekten Kamera- und Softwareeinstellungen herausgefunden haben, funktioniert die allgemeine Erkennung der Berührungen recht gut. Wir lieben es regelrecht die grünen Punkte, welche so schön an unseren Fingerspitzen haften, über den Tisch zu jagen.

We have received all the hardware by now, including an incredible wide-angle lens for our camera and the new beamer, and have started our testing sessions for the best setup. With some hints in postings in the NUI Group Forums and a couple of days of testing different materials, we have built a compliant surface that works very well with zero force, while still providing a good projection image. We are about to build the construction for the mobile prototype based on our research. Gregor already made some new images that will be on his blog soon!

On Gregor’s blog there are details (including great pictures!) of our “silicone improvement” we did today – which means, we added a 0,5mm thin silicone layer to our acrylic plate to improve blob detection and haptic experience.

As our multitouch table project at Strukt grows there’s more and more crafty action going on in our office. Today we applied a silicone layer to the acrylic glass plate to improve the touchsensibility of the surface. But through the last days there was also a lot of drilling and soldering and a total of 198 LEDs put into our frame as well. Also the demo application is already done, the new wideangle lens for the camera arrived to day and the new projector is about to be delivered. So over the next days there are definitely more updates on the way.

Unser Multitouch Projekt bei Strukt wächst und gedeiht und mit einher geht ein reges handwerkliches Schaffen im Büro. Heute haben wir eine Silikonschicht auf die Acrylplatte aufgetragen um die Berührungsempfindlichkeit zu steigern. Aber auch während der letzten Tage wurde fleißig gebohrt, gelötet und insgesamt 198 LEDs verbaut. Außerdem wurde die Demoapplikation fertiggestellt, das neue Weitwinkelobjektiv für die Kamera geliefert und auch der neue Beamer ist bald zu erwarten. Also die nächsten Tage wird sicher viel weitergehen, somit auch mehr Updates kommen.

Safety First.

Silicon is a funny acting substance.

After pouring the silicon on the plate ….

… it has to be spread evenly.

Finally you cover it loosely and bake it at room temperature for 15 hours.

At our recently founded company Strukt we have decided to form a research team to build a multitouch table.

Based on my research and our technical director Thomas’ knowledge and experience in similar projects he realised while working at meso, we are about to finish the prototype. We have done testing with various hardware: camera, optics, beamers, infrared leds, acrylic tops with different compliant surfaces and diffuser screens. Have a look at Gregor’s documentation pictures, although they are already one week old – in the meantime there is much more to see (more photos in one of my next posts). Finally we have found good solutions for all challenges, and have a table that measures 150×90 cm, which will most probably be the final size. Also the software part, all realised with vvvv, is in beta-stage and blob recognition is already working. Next will be the building of a proper case for our prototype and development of basic applications and frameworks to develop applications in vvvv flash.