Learning Vortex Dynamics for Fluid
Inference and Prediction

ICLR 2023

Yitong Deng1,2, Hong-Xing "Koven" Yu2, Jiajun Wu2, and Bo Zhu1

1 Dartmouth College; 2 Stanford University
Abstract

We propose a novel differentiable vortex particle (DVP) method to infer and predict fluid dynamics from a single video. Lying at its core is a particle-based latent space to encapsulate the hidden, Lagrangian vortical evolution underpinning the observable, Eulerian flow phenomena. Our differentiable vortex particles are coupled with a learnable, vortex-to-velocity dynamics mapping to effectively capture the complex flow features in a physically-constrained, low-dimensional space. This representation facilitates the learning of a fluid simulator tailored to the input video that can deliver robust, long-term future predictions. The value of our method is twofold: first, our learned simulator enables the inference of hidden physics quantities (e.g., velocity field) purely from visual observation; secondly, it also supports future prediction, constructing the input video's sequel along with its future dynamics evolution. We compare our method with a range of existing methods on both synthetic and real-world videos, demonstrating improved reconstruction quality, visual plausibility, and physical integrity.

Paper

[PDF]
Code

[Github]
Video
Citation
@inproceedings{deng2023vortex,
 title={Learning Vortex Dynamics for Fluid Inference and Prediction},
 author={Yitong Deng and Hong-Xing Yu and Jiajun Wu and Bo Zhu},
 booktitle={Proceedings of the International Conference on Learning Representations},
 year={2023},
}