Hello David,
You could track a red led (or bright dot) in GridFlow by using [lti.SplitImageToRGB] and sending the red channel to [#motion_detect]. It outputs x and y values. I have some images of drawings made after this here : http://artengine.ca/acastonguay/projets/plotter/. You'll also find some images of early motion detection studies by the Galbraiths that puts this in perspective.
Cheers,
Alexandre
Le Vendredi 6 Octobre 2006 18:27, Mathieu Bouchard a écrit :
On Fri, 6 Oct 2006, David Powers wrote:
Finally, are there more advanced motion detection strategies? Like, say, could I somehow put a bright red dot on my index finger, and just detect where my index finger is? I'd love to hear what people have done.
Alexandre Castonguay has done that a few years ago using high-intensity LED. I'm doing the same for my girlfriend's project, which will be presented as Café Esperanza on the 19th at 20:00.
Café Esperanza is at the corner of Saint-Viateur & Saint-Laurent, in the city mentioned below:
_ _ __ ___ _____ ________ _____________ _____________________ ...
| Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Is there some easy installation to try out GridFlow? How hard is it to install the Fedora Core and add the planet ccrma packages, for a complete Linux beginner?
I'm on WinXP, but I have tried out PD in Linux using Pure:Dyne. The gridflow examples didn't seem to work, however... Nor did PMPD, if I remember. They just, well, didn't do anything...
I also don't know, how hard it would be to get the Logitech camera running under Linux. And to get my wireless USB running in Linux, which is the only way I can get on the internet, to download the drivers and software I might need to make everything run...
~David
On 10/6/06, Alexandre Castonguay acastonguay@artengine.ca wrote:
Hello David,
You could track a red led (or bright dot) in GridFlow by using [lti.SplitImageToRGB] and sending the red channel to [#motion_detect]. It outputs x and y values. I have some images of drawings made after this here : http://artengine.ca/acastonguay/projets/plotter/. You'll also find some images of early motion detection studies by the Galbraiths that puts this in perspective.
Cheers,
Alexandre
Le Vendredi 6 Octobre 2006 18:27, Mathieu Bouchard a écrit :
On Fri, 6 Oct 2006, David Powers wrote:
Finally, are there more advanced motion detection strategies? Like, say, could I somehow put a bright red dot on my index finger, and just detect where my index finger is? I'd love to hear what people have done.
Alexandre Castonguay has done that a few years ago using high-intensity LED. I'm doing the same for my girlfriend's project, which will be presented as Café Esperanza on the 19th at 20:00.
Café Esperanza is at the corner of Saint-Viateur & Saint-Laurent, in the city mentioned below:
_ _ __ ___ _____ ________ _____________ _____________________ ...
| Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Hi again,
You have to get the cvs version for GridFlow to use the lti objects. You can find the info here : http://gridflow.ca. I have a logitech camera (quickcam 4000) that is supported by http://saillard.org. For the usb wireless adaptor I am not sure.
Once your network access is configured, you can get a lot of help to install and run pd (+ GridFlow) from irc.freenode.net, channel #dataflow.
Good luck,
Alexandre
Is there some easy installation to try out GridFlow? How hard is it to install the Fedora Core and add the planet ccrma packages, for a complete Linux beginner?
I'm on WinXP, but I have tried out PD in Linux using Pure:Dyne. The gridflow examples didn't seem to work, however... Nor did PMPD, if I remember. They just, well, didn't do anything...
I also don't know, how hard it would be to get the Logitech camera running under Linux. And to get my wireless USB running in Linux, which is the only way I can get on the internet, to download the drivers and software I might need to make everything run...
~David
On 10/6/06, Alexandre Castonguay acastonguay@artengine.ca wrote:
Hello David,
You could track a red led (or bright dot) in GridFlow by using [lti.SplitImageToRGB] and sending the red channel to [#motion_detect]. It outputs x and y values. I have some images of drawings made after this here : http://artengine.ca/acastonguay/projets/plotter/. You'll also find some images of early motion detection studies by the Galbraiths that puts this in perspective.
Cheers,
Alexandre
Le Vendredi 6 Octobre 2006 18:27, Mathieu Bouchard a écrit :
On Fri, 6 Oct 2006, David Powers wrote:
Finally, are there more advanced motion detection strategies? Like,
say,
could I somehow put a bright red dot on my index finger, and just
detect
where my index finger is? I'd love to hear what people have done.
Alexandre Castonguay has done that a few years ago using
high-intensity
LED. I'm doing the same for my girlfriend's project, which will be presented as Café Esperanza on the 19th at 20:00.
Café Esperanza is at the corner of Saint-Viateur & Saint-Laurent, in
the
city mentioned below:
_ _ __ ___ _____ ________ _____________ _____________________ ...
| Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
David Powers schrieb:
Is there some easy installation to try out GridFlow? How hard is it to install the Fedora Core and add the planet ccrma packages, for a complete Linux beginner?
When I used Fedora Core about one year ago I was happy with it, but in the meantime I think the development is stopped(?), so Ubuntu might be a better choice, though I have not tested that by myself, I am on a Debian system (which ubuntu is also based on). this might be helpful: http://puredata.info/docs/developer/ http://puredata.info/docs/developer/Debian
marius.
On Fri, 6 Oct 2006, Alexandre Castonguay wrote:
You could track a red led (or bright dot) in GridFlow by using [lti.SplitImageToRGB] and sending the red channel to [#motion_detect].
[#inner (3 1 # 1 -1 -1)] is better because it will filter out white, purple and yellow, while just splitting channels doesn't.
[#inner (3 1 # 1 0 0)] might be a faster way of getting the red channel, but I haven't tried. For extracting all three channels at once, [lti.SplitImageToRGB] is surely faster than three #inners at once.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
On Fri, 6 Oct 2006, Mathieu Bouchard wrote:
[#inner (3 1 # 1 0 0)] might be a faster way of getting the red channel, but I haven't tried.
I mean that I haven't checked whether it's faster than the lti splitter; I've always extracted channels using #inner.
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
-
Alexandre Castonguay -
David Powers -
Marius Schebella -
Mathieu Bouchard