A Multi-Pass GAN for Fluid Flow Super-Resolution

Abstract

We propose a novel method to up-sample volumetric functions with generative neural networks using several orthogonal passes. Our method decomposes generative problems on Cartesian field functions into multiple smaller sub-problems that can be learned more efficiently. Specifically, we utilize two separate generative adversarial networks: the first one up-scales slices which are parallel to the XY-plane, whereas the second one refines the whole volume along the Z-axis working on slices in the YZ-plane. In this way, we obtain full coverage for the 3D target function and can leverage spatio-temporal supervision with a set of discriminators. Additionally, we demonstrate that our method can be combined with curriculum learning and progressive growing approaches. We arrive at a first method that can up-sample volumes by a factor of eight along each dimension, i.e., increasing the number of degrees of freedom by 512. Large volumetric up-scaling factors such as this one have previously not been attainable as the required number of weights in the neural networks renders adversarial training runs prohibitively difficult. We demonstrate the generality of our trained networks with a series of comparisons to previous work, a variety of complex 3D results, and an analysis of the resulting performance.

SCA 2019, further details at: https://ge.in.tum.de/publications/2019-multi-pass-gan/

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Conditional Generative Adversarial Nets
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Accelerating Eulerian Fluid Simulation With Convolutional Networks
• PDE-Net: Learning PDEs from Data
• Artistic style transfer for videos
• Least Squares Generative Adversarial Networks
• tempoGAN: A Temporally Coherent, Volumetric GAN for Super-resolution Fluid Flow
• Data-Driven Synthesis of Smoke Flows with CNN-based Feature Descriptors
• Deep Learning the Physics of Transport Phenomena
• Deep Scattering: Rendering Atmospheric Clouds with Radiance-Predicting Neural Networks

Cited by in corpus (9)

• Unsupervised deep learning for super-resolution reconstruction of turbulence
• Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels
• From coarse wall measurements to turbulent velocity fields through deep learning
• Generative Modeling of Turbulence
• A neural network multigrid solver for the Navier-Stokes equations
• Fluid Simulation on Neural Flow Maps
• Reconstruction of three-dimensional turbulent flow structures using surface measurements for free-surface flows based on a convolutional neural network
• Neural UpFlow: A Scene Flow Learning Approach to Increase the Apparent Resolution of Particle-Based Liquids
• Dynamic Upsampling of Smoke through Dictionary-based Learning