How to setup a VVVV to Processing via OSC testing environment- UNDER REVISION

a) “Right Click” on Video In on/off button (middle square) Now you should see video stream from directshow videoIn - Note the source and format of
video input should be configured dependant of your camera.

b) To turn on blobtracking “Right Click” on “blocktracking on/off” square (now you see detected blobs)

c) You can change Threshold by “Right Clicking” on Threshold bar and dragging it Up and Down (just play with it, to see what effect it does)


Now VVVV is searching for blobs and sending OSC messages about detected blobs to localhost 9000 by UDP. Leave VVVV opened.

a) Start processing

b) Open processing sketch by pressing Ctrl+O , and select the file from VVVV_processing_blobtracker.zip named: vvvv_blobtracker/vvvv_blobtracker.pde

c) Click Run on quick bar. (if it errors on “framerate” please replace with “frameRate")

d) Drawing detected blobs function is in Tracker.java on line 118, function public void draw(PGraphics g)

Thanks alot Nick, Great information to have, I am in the process of writing a full tutorial on VVVV, OSC, to Processing methods of this because I think it will be alot more robust as well as the open source approach to development. Look forward to seeing it next week with some example processing apps.

I hope I get some free time on the weekend

no problem man, sweet, sounds good, looking forward to it

for now me and my crew will stick with java and processing, because we found VVVV a little hard to work with for the moment (we’re not really programmers ourselves, more interaction designers)
we’re also looking into the reactivision software, look on what we can use....

we hope to get 2 full time programmers (students) somewhere in february to help us start developing stuff

if anything interesting pops up i’ll let you know

we played around with reactivision software today, which has some nice features

Yea Grav great find on that reactable software, just so others know i am going to post site here..

“eacTIVision is an open source, cross-platform computer vision framework for the fast and robust tracking of fiducial markers attached onto physical objects, as well as for multi-touch finger tracking. It was mainly designed as a toolkit for the rapid development of table-based tangible user interfaces (TUI) and multi-touch interactive surfaces. This framework has been developed by Martin Kaltenbrunner and Ross Bencina at the Music Technology Group at the Universitat Pompeu Fabra in Barcelona, Spain as part of the the reacTable* project, a novel electro-acoustic music instrument with a tangible user interface.”

Get it here: http://mtg.upf.edu/reactable/?software

Thank tj_joobs he pasted the URL in irc
I had allready visited that site once, but probably to quick because i didnt notice anything about software or open source :$

Hey guys,
three weeks ago I found out about vvvv, in my current project I use this software solely for the multi touch purpose (so without using processing or any other.) I’ve created a patch that captures video and by turning on or off the clear pin on the renderer you can easily draw stuff. I will upload a video to youtube soon.

[edit]Oh I also want to remark that the reactivision framework is also present in vvvv as a freeframe plugin. I use it for tracking coasters on my bartable.

k sounds cool, let us know when the video is on youtube

any chance of you being able to upload that vvvv script when our blog/site is online?

Right now the patch is a big mess where all the functions are scattered over the patch.

I’m working on a patch where everything is explained. I will upload when I have time for that, right now I need to finish the report about the prototype.

I’ve just uploaded a video to youtube. It shows me testing the attractor function in vvvv on an earlier version of my prototype. It’s bit buggy and laggy. But I think it’s because I use christmas lighting as an IR source and a sweex camera which seems to operate max at 5 fps because it thinks the lighting condition is too low…

http://www.youtube.com/watch?v=jZGcVTwRV-w

Was that all done in vvvv? Looks quite cool even though it has a poor frame rate.

Is there any kind of camera distortion/projector calibration?

looks nice hype, good luck with your report, looking forward to see your script

Everything is done in vvvv, vvvv contains freeframe plugins which provide the contour, blob, motion, reactivision and other functions. But it has also functionality to send via OSC, since I have no knowledge about what to do with that I stayed patching within the vvvv.

There is lots of camera distortion and as yet I haven’t programmed a way to undo that. Do only way to counter the effects were to position the camera at the right spot and use mapping to calibrate the contours.

I will post a way cooler video today later.

Looks great hype, I myself am trying to work out the patches in VVVV, and am defiantly find it messy like you say.

Thanks for the video, hope to see more

Hi Jens, thanks alot for posting, this should really help out the people over here
Credit for the detailed description goes to Nikolajus though
Anyway, thanks again for dropping by

Hey Jens, Thank you sooooooo much for your works, I cannot wait until I get it up and running myself, a lot of computer issues here on my main PC my directshow is toast.

I am trying to rewrite the above tutorial, Most of the links did not work properly, so I am revising it with correct links and making it easier to follow.

Please let me know if you would like to make any changes to the tutorial to make it simpler/correct, I myself am going to give the option of downloading one big ZIP file including everything one would need such as; directx 9.0c, javaOSC, VVVV, Processing, and of course Jens beautiful v4p and pde files, organized nicely in one folder. I hope this will make things even easier for everyone.

Regards,
Chris

Hey, I used vvvv and osc for my project for uni (http://www.andybest.net/vids/2007inst2.html) and I had some (very) simple calibration code as you can see in the video- it doesn’t calibrate for angular distortion, but its a good starting point i suppose… i’d be happy to share it if anyone is interested, although, as i said, it could use much improvement.

The code is in Lingo, as I used Macromedia Director to produce the visuals for it, but it should be easy to port over to any other language that you want.

in case anyone was wondering, i’m tracking an infrared LED attatched to the paintbrush, and it has a switch on it that makes it draw.

Thanks also to Jens- you really helped me out a while ago on your blog when i was researching.

Hey Goobers,
I saw your video and it looked nice . Guess that panel works without FTIR, or did it? I think that if you use FTIR you can put a piece of siliconrubber on the brush and paint more accurate.