Funny, I thought of the affine transform too. I was going to have the visitor wave their hands in front of the camera to create bounding trapezoids (size determined by the size of whatever you're waving, and persistence determined by how long you kept waving near the same spot; they would disappear some time after being created). Then to create the actual image, each trapezoid contains the user's image, suitably faded and alpha-blended to maintain contrast with a uniform overall brightness. Mike, to make Barnsley's fern, two of the people would have to be very small (infants would do) and one would have to be even smaller and extremely thin (perhaps a pet garter snake trained to stay still :-) Or you could abandon the "lying on the ground" idea, and instead place the camera at waist level and allow people to change their size by walking towards and away -- but then the placement of the smaller bounding boxes is constrained roughly to a horizon line -- or else the museum has to contain a jungle gym or similar climbable 3-D lattice On Tue, Oct 19, 2010 at 12:00, Mike Stay <metaweta@gmail.com> wrote:
There are a bazillion things you can do with webcam and a projector. One I just thought of is taking bounding boxes around people laying on the floor as an affine transformation and then generating an IFS. Three people will give you some variant of the Serpinski triangle; four properly arranged will give you Barnsley's fern. [...]
-- Robert Munafo -- mrob.com Follow me at: mrob27.wordpress.com - twitter.com/mrob_27 - youtube.com/user/mrob143 - rilybot.blogspot.com