Post-processing framework for enhancing liquid surfaces in VFX pipeline
Jong-Hyun Kim* Jaeho Im*** Jung Lee***
(* : Kangnam University, ** : Engine Visualwave Co., *** : Hallym University)
IEEE Access 2021
Jong-Hyun Kim* Jaeho Im*** Jung Lee***
(* : Kangnam University, ** : Engine Visualwave Co., *** : Hallym University)
IEEE Access 2021
Abstract : Physically-based simulation is being expanded to simulate various materials such as deformable bodies, sand, and lava as well as liquid, and has been widely used as plug-in and software in actual related industries. Contrary to this trend, however, most of the simulation software products are built as an in-house solution, making it difficult for users to add new features not provided by the software. In this paper, we propose a post-processing framework that can improve the visual quality of liquid surfaces by receiving only the positions of liquid particles created in the existing simulation tools. Our framework includes the following methods: 1) analyzing liquid flow from particle position, 2) sampling technique for creating liquid sheets. Our framework improves the detail of liquid surfaces by temporarily adding new liquid particles through sampling of particles from existing simulation techniques. Our technique does not depend on any particular simulation tool. Experiments with various scenarios have also shown that liquid sheet detail, which was difficult to express in traditional software, has been improved, and problems such as break-up fluid surfaces and noisy surfaces have been mitigated. In addition, our research is a post-processing framework that uses only particle position as input data, making it easy to integrate with existing simulation tools that provide particle position.
[paper]