Title: Fluid Animation Based on Statistical Method
Abstract
One of the most intriguing problems in computer graphics is the
simulation of fluid-like behavior. Most researches in fluid simulation
have been focused on techniques for solving the famous incompressible
Navier-Stokes equations for fluid flow to obtain physically accurate
simulations. In this presentation, we introduce some fundamental
concepts first, and then propose a novel method for fluid animation. We
aim to obtain visually plausible fluid animation by constructing from
the results of physically accurate simulations. We synthesize fluid
motion from simulation samples precomputed by a full numerical solver.
We dice simulation sample data in order to capture the properties of
local fluid behavior and apply statistical techniques to abstract the
dominant characteristics from each sample fragment so that these
fragments can be selected and rearranged fast. Our technique will be
suitable to be applied to visual effects design in movies as well as
applications such as video games and virtual environments.