I first read about this on Bruce Schneier’s blog, believe it or not. Pretty cool.
One Response to “Dice-o-matic”
Pretty awesome. I think you could generalize this; You’ve seen the trackballs with little dots all over them? The dots are a unique pattern so that any small sample of a few dots will indicate the position on the sphere by their relative orientation. If you did this with trackballs, you could have a LOT more unique values, and wouldn’t have to worry about orientation as much.
The processing required would be much greater (I’m not sure you could write the scanner in .NET at this point) but you could get floating-point values by calculating the longitude and latitude of the center of the ball as it faces the camera. If the dots were printed in Infrared ink and the camera had a lens filter, it should become very easily manageable in software.
One Response to “Dice-o-matic”
Pretty awesome. I think you could generalize this; You’ve seen the trackballs with little dots all over them? The dots are a unique pattern so that any small sample of a few dots will indicate the position on the sphere by their relative orientation. If you did this with trackballs, you could have a LOT more unique values, and wouldn’t have to worry about orientation as much.
The processing required would be much greater (I’m not sure you could write the scanner in .NET at this point) but you could get floating-point values by calculating the longitude and latitude of the center of the ball as it faces the camera. If the dots were printed in Infrared ink and the camera had a lens filter, it should become very easily manageable in software.
By jYopp on May 31, 2009 at 12:46 pm