Counting Particles: a simple and fast surface reconstruction method for particle-based fluids

We present a novel and efficient surface reconstruction for particle-based fluid methods. Although particle-based methods are practical for computing level-sets that represent liquid interfaces, these methods are computationally expensive when the number of particles increases considerably due to the intense usage of particle approximations. This paper introduces a simple level-set approximation using a discrete indicator function (DIF) defined by counting particles inside grid cells. Our method is fast, easy to code, and can be adapted straightforwardly in particle-based solvers, even implemented in GPU. Moreover, we show the effectiveness of our approach through a set of experiments against prior surface reconstruction methods.