MODELING DYNAMIC SURFACES WITH OCTREES

Past uses of octrees have been for representation of static objects. We discuss extensions necessary to model dynamic surfaces. One particularly important aspect of this is the ability to represent expanding surfaces that grow to be arbitrarily large. Our enhanced octree does exactly this, and models contraction as well. The ability to represent dynamic surfaces allows us to apply octrees to new problems which could not previously have been modeled with static octrees. One such problem is the entropy of random surfaces. Using dynamic octrees, we produced a simulation of self-avoiding random surfaces using Monte Carlo techniques.