What I’m trying to do is this in vvvv.
I’ve tried contour tracking; color tracking; a red filter on the camera, (laser pointer is red) > a threshold pixel shader > contour tracking. That last one seems to be working pretty well, but is contour tracking overkill for this? One problem is it getting messed up by seeing what it’s projecting. (actually haven’t tried it with a projector yet, but a second monitor)
One other thing, tho it’s less important to me right now, is being able to calibrate the camera as in the screenshot on that page, where you select where in the camera’s view the projection is. If I could map x AND y values to other x and y values, at the same time, (e.g. map .6,.8 to 1,1) then I’d be set. I hope that’s easy to understand. Any ideas?
One other thing, tho it’s less important to me right now, is being able to calibrate the camera as in the screenshot on that page, where you select where in the camera’s view the projection is. If I could map x AND y values to other x and y values, at the same time, (e.g. map .6,.8 to 1,1) then I’d be set. I hope that’s easy to understand. Any ideas?
look at the following nodes: Map (Value) or its sisters, ApplyTransform (Transform) and the other transform nodes… in fact, a transform node is a whole new coordinate space, if you map this space, simply with a quad as indicator, you just have to use the transform of this quad and use ApplyTransform (Transform) to map a -0.5…0.5 space (the quad default size) onto the region where you see the quad.
Yeah, I know Map, and I just checked out ApplyTransform, but what I’m wanting to do is, have the camera be able to see more than the range of the projector (like in the pic on that page) and select in the view of the camera where the range of the projector is, then map any laser pointer movement in that range to the -1,-1…1,1 range so that the laser pointer’s movement and what’s shown on the projector match up. See what I’m saying? If you don’t, let me know…
