Reconstructing surfaces of particle-based fluids using anisotropic kernels

Abstract

In this article we present a novel surface reconstruction method for particle-based fluid simulators such as Smoothed Particle Hydrodynamics. In particle-based simulations, fluid surfaces are usually defined as a level set of an implicit function. We formulate the implicit function as a sum of anisotropic smoothing kernels, and the direction of anisotropy at a particle is determined by performing Principal Component Analysis (PCA) over the neighboring particles. In addition, we perform a smoothing step that repositions the centers of these smoothing kernels. Since these anisotropic smoothing kernels capture the local particle distributions more accurately, our method has advantages over existing methods in representing smooth surfaces, thin streams, and sharp features of fluids. Our method is fast, easy to implement, and our results demonstrate a significant improvement in the quality of reconstructed surfaces as compared to existing methods.

An Adaptive Model for Particle Fluid Surface Reconstruction

“…On contrast, our method protects these (d) show the effect of almost calm at the same frame. We notice that our method produces the same smoothness as [10], and our method is without the process of smoothing. Figure 5 shows an animation of flowing water on a stair scene with 600 K particles.…”

“…Figure 4 shows the water splashes into a pool scene with 400 K particles. It shows a comparison of the method of [10] and our adaptive method. The left is rendered with our method, such as (a) and (c).…”

“…The left is rendered with our method, such as (a) and (c). The right is rendered with the method of [10], such as (b) and (d). (a) and (b) show the effect of fluctuations at the same frame.…”

“…(a) and (b) show the effect of fluctuations at the same frame. Because the method of [10] uses a processing of smoothing step, we can see that the effects of ripples disappear and some details in thin particles are also lost. On contrast, our method protects these (d) show the effect of almost calm at the same frame.…”

“…Our method is inspired by the work of [10], which replaced the spherical kernels using ellipsoidal kernels, and had better accuracy than standard SPH. However, some defects still exist.…”