First off I am a bit of a newbi here, but I’ve read thru the Multi-touch books and gotten the hardware design stuff figured out (working out the final details on a 42” DSI setup).. So now I am turning to the software stack.. Here is where I am getting a bit confused.. Specificly what I am looking to do is build something that works with the MS Multi-touch framework inside windows 7 and also supports tags.. Given that Fiducials only support 217 symbles I am actually looking to roll my own or port some C# code that handles QR codes.. ]
This brings me to my two questions..
1) What does the software stack really look like as far as video stream->[app1]->[interface/protocal]->[app2]-> mouse event.. (I’ve looked all over the place but a good sequence diagram or even high level architectual overview is all but unfindable (by me)… From what I can tell (with Multitouchvista) its video stream ->TUIO provider -> TUIO Protocal -> Multitouchvista->Mouse event. With TUIO Protocal acting as a physical/process boundry between the provider and Multitouchvista, but its all still very fuzzy to me..
2) If one was going to roll his own Tag finder/tracker where in the stack would it go? can it be done in the Managed side of things or should it be handled by the touch provider?
Thanks in advance