mindnut

I'm attempting to develop an interactive that needs to recognize a shape in a live video stream. Upon seeing that shape it will trigger a response. Is it necessary to use a Kinect or can I get away with a usb camera? Can anyone suggest the Jitter objects I should be looking at? Thank you

I'm attempting to develop an interactive that needs to recognize a shape in a live video stream.  Upon seeing that shape it will trigger a response.  Is it necessary to use a Kinect or can I get away with a usb camera?  Can anyone suggest the Jitter objects I should be looking at?  Thank you


to-kinect-or-not-to-kinect

You'll probably need to provide more details for a good answer. Is it a shape composed using the human body? If so, probably Kinect is the way to go. If not, you might have better luck doing something more like "blob tracking" which cv.jit can possibly help you with. Keep in mind that some kinds of CV object recognition can be extremely difficult to implement: 

You'll probably need to provide more details for a good answer. Is it a shape composed using the human body? If so, probably Kinect is the way to go. If not, you might have better luck doing something more like "blob tracking" which cv.jit can possibly help you with. Keep in mind that some kinds of CV object recognition can be extremely difficult to implement: http://xkcd.com/1425/ ;)


To Kinect or not to Kinect