relatively new to jitter…im trying to create a system where with live camrea input i can detect certain shapes/tags on the bottom of blocks, and therefore get accurate location information for each individual block. i have got a system working using different colours, but i think for the project it may work alot better if i could recognise some form of shape instead.
can anyone point me in the right direction to getting this working..is it even a possibility?
thanks very much.
look in the list archives.
this has come up repeatedly.
cv.jit objects or softVNS will help you with this.
v a d e //
I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I
LIVE! I LIVE! I LIVE! I LIVE!
You will not be saved by the Holy Ghost. You will not be saved by the
In fact, YOU WILL NOT BE SAVED!
thanks very much…ill check the list archives out.
You might find ARToolKit of interest, recognizing symbols on blocks is
what you’re after:
and there are a number of other open source projects along these
lines. Another resource that might be of interest is Martin
Kaltenbrunner’s page on tangible musical interfaces, several of which
(including the reacTable) use shape recognition.
—– |(*,+,#,=)(#,=,*,+)(=,#,+,*)(+,*,=,#)| —–
thanks for the help so far…just another question..i have been playing around with all your ideas for the recognition but with all accept using color recognition as soon as the object leaves the camrea view it loses it and cannot track it when it returns (obviously)….does anyone know of any way or an other system of doing this that could enable me to track objects that will move out of the camera view?