Data-Driven Synthesis of Smoke Flows with CNN-based Feature Descriptors
arXiv:1705.01425 · doi:10.1145/3072959.3073643
Abstract
We present a novel data-driven algorithm to synthesize high-resolution flow simulations with reusable repositories of space-time flow data. In our work, we employ a descriptor learning approach to encode the similarity between fluid regions with differences in resolution and numerical viscosity. We use convolutional neural networks to generate the descriptors from fluid data such as smoke density and flow velocity. At the same time, we present a deformation limiting patch advection method which allows us to robustly track deformable fluid regions. With the help of this patch advection, we generate stable space-time data sets from detailed fluids for our repositories. We can then use our learned descriptors to quickly localize a suitable data set when running a new simulation. This makes our approach very efficient, and resolution independent. We will demonstrate with several examples that our method yields volumes with very high effective resolutions, and non-dissipative small scale details that naturally integrate into the motions of the underlying flow.
14 pages, 17 figures, to appear at SIGGRAPH 2017, v2 only fixes small typos
Cited by in corpus (11)
- Deep Fluids: A Generative Network for Parameterized Fluid Simulations
- Deep Learning in Deterministic Computational Mechanics
- FEA-Net: A Physics-guided Data-driven Model for Efficient Mechanical Response Prediction
- Deep Scattering: Rendering Atmospheric Clouds with Radiance-Predicting Neural Networks
- A Multi-Pass GAN for Fluid Flow Super-Resolution
- ScalarFlow: A Large-Scale Volumetric Data Set of Real-world Scalar Transport Flows for Computer Animation and Machine Learning
- DeepWarp: DNN-based Nonlinear Deformation
- Generative Modeling of Turbulence
- Lagrangian Neural Style Transfer for Fluids
- Neural UpFlow: A Scene Flow Learning Approach to Increase the Apparent Resolution of Particle-Based Liquids
- Multi-Grid Graph Neural Networks with Self-Attention for Computational Mechanics