Hi Paul This would lead to rather large bounding boxes wouldn't it? From the pre-transformed bounding box you would have to take the circumscribed sphere (the sphere through the corners) and for the resulting you would have to use the bounding box containing this sphere. Then the original box is contained in the sphere and the sphere is contained in the resulting box. In effect every transform would increase the sphere with sqrt(3). Or is there a smart trick that I didn't think about?
Bests, Nico Kruithof On Thu, May 14, 2009 at 9:32 PM, Paul Martz <[email protected]> wrote: >> The fast method would be to compute the eight corners, transform them, and > create >> a new BB that contains the eight transformed points. > > I guess, if you don't care about accuracy, then a more optimized approach > would be to transform the center of the bounding box and its "radius" and > then compute a new BB from the transformed data. (The OP did ask about the > most efficient method, IIRC.) > -Paul > > _______________________________________________ > osg-users mailing list > [email protected] > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
