Data-Driven Synthesis of Smoke Flows with CNN-based Feature Descriptors

摘要

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 u001efluid data such as smoke density and u001eflow velocity. At the same time, we present a deformation limiting patch advection method which allows us to robustly track deformable u001efluid regions. With the help of this patch advection, we generate stable space-time data sets from detailed u001efluids 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 effiu001ccient, and resolution independent. We will demonstrate with several examples that our method yields volumes with very high effu001dective resolutions, and non-dissipative small scale details that naturally integrate into the motions of the underlying u001eflow.