Multitouch Patch

First post of a series…
Hi all.
I’m trying to build a multitouch surface using all the info found on various site (FTIR). My first question is how to “connect” the blobs recognise by the cam using the patch found on
I’m an italian “lamer” and i need to better understand how to…
Thanks in advance

hi headhole,

direct beyond the video you linked is a picture of the patch.
so i don’t get a clue about your problem.

OK, let me try to explain what i want to do: the patch actually record the blob position and “write” on the screen (because of the “400” frames keeped in memory) something similar to a paint mesh. But i want to reduce the numbers of frames from 400 to 10-20 and connect the blobs with lines. Is that simple to do?

yes… no text …

…and so? any avdice on what i have to add? the better way is to post a screenshot of what i have to add due to connect with lines the blobs.

Hehe… well, Kalle forgot to tell that the word simple is relative offcours.

The Queue node has a ‘do insert’ pin, yours is set to 1 at the moment, meaning it wil add a new value to your ‘buffer’ every vvvv frame.

What you want to do is add a new value only every X-frames, or X-sec, wich is easier. (use and LFO and Change node)

Now you got some points that are collected in a steady time interval, and you can use the Rope or Line node to draw your line.

callmenames.v4p (11.9 kB)

because you have more than one blob you have to take care about the BinSize.

headhole.v4p (13.7 kB)

Thank you kalle & West. I’ll try the patches posted asap and will your advices about the… BinSize/LFO?

Ok, I’ve made some test using the patches. Now i’m starting to understand how to use the vvvv for my project. I’m here another time because stopped by a little problem: I was trying to print near the blob the position X and Y (and maybe something other infos)… can’t understand what I’ve have to use (and maybe why?).
Thanks in advance.

I’m getting crazy.
I’m not able to write the X and Y positions (coordinates) of the blobs detected on the screen! Probably because I can’t understand how correctly use the TextEx9 and the “X Mesh”: probably because i don’t know where or what’s an X Mesh? Or else because TextEx9 is not the better solution?

Bene (well).
I’ve got something using Text (GDI) and converting the X and Y numbers with AsString, joining toghether with AddString.
But the rendering is in the GDI window. I need the rendering in the DirectX Render (RendererEX9). I know it’s possible and should be easy (as kalle sad on my previous post) but i need help.

callmenames12_2007.08.30-11.06.10.jpg (112.4 kB)

GDITexture (EX9.Texture) has a helpfile.

As always kalle you are my angel.
Now I’m able to show the position of the donuts on the screen. But please, make me understand why i have to Scale the render?
For today i want to connect all the point toghether, but not with a single line: every point must connect to the other points on the screen. Any advice?

callmenames16.v4p (48.7 kB)

try the node ConnectAll :)

ConnectAll (2d)