Particle-Based Fluid Surface Rendering with Neural Networks

Surface reconstruction for particle-based fluid simulation is a computational challenge on par with the simula- tion itself. In real-time applications, splatting-style rendering approaches based forward of particle impostors are prevalent, but they suffer from noticeable artifacts. this paper, we present technique that combines simulated features deep-learning image manipulation to improve quality be perceptually similar ray tracing solutions, circumventing cost, complexity, and limitations exact surface by replacing it flat cost neural network pass. Our solution idea training generative deep networks pairs consisting cheap impostor renders ground truth high ray-traced images.