Data‐driven projection method in fluid simulation

Yang, Cheng; Yang, Xubo; Xiao, Xiangyun

doi:10.1002/cav.1695

Cited by 104 publications

(75 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other graphics works have targeted learning flow descriptors with CNNs [CT17], or learning the statistics of splash formation [UHT17], and two‐dimensional control problems [MTP∗18]. While pressure projection algorithms with CNNs [TSSP16, YYX16] shares similarities with our work on first sight, they are largely orthogonal. Instead of targeting divergence freeness for a single instance in time, our work aims for learning its temporal evolution over the course of many time steps.…”

Section: Related Work and Backgroundmentioning

confidence: 96%

Latent Space Physics: Towards Learning the Temporal Evolution of Fluid Flow

Wiewel

Becher

Thuerey

2019

Computer Graphics Forum

204

172

View full text Add to dashboard Cite

We propose a method for the data‐driven inference of temporal evolutions of physical functions with deep learning. More specifically, we target fluid flow problems, and we propose a novel LSTM‐based approach to predict the changes of the pressure field over time. The central challenge in this context is the high dimensionality of Eulerian space‐time data sets. We demonstrate for the first time that dense 3D+time functions of physics system can be predicted within the latent spaces of neural networks, and we arrive at a neural‐network based simulation algorithm with significant practical speed‐ups. We highlight the capabilities of our method with a series of complex liquid simulations, and with a set of single‐phase buoyancy simulations. With a set of trained networks, our method is more than two orders of magnitudes faster than a traditional pressure solver. Additionally, we present and discuss a series of detailed evaluations for the different components of our algorithm.

show abstract

Section: Related Work and Backgroundmentioning

confidence: 96%

Latent Space Physics: Towards Learning the Temporal Evolution of Fluid Flow

Wiewel

Becher

Thuerey

2019

Computer Graphics Forum

204

172

View full text Add to dashboard Cite

show abstract

“…An LSTM‐based method for predicting changes of the pressure field for multiple subsequent time steps has been presented by [WBT18], resulting in significant speed‐ups of the pressure solver. For a single time step, a CNN‐based pressure projection was likewise proposed [TSSP17, YYX16]. In contrast to our work, these models only replace the pressure projection stage of the solver, and hence are specifically designed to accelerate the enforcement of divergence‐freeness.…”

Section: Related Workmentioning

confidence: 99%

Deep Fluids: A Generative Network for Parameterized Fluid Simulations

Kim

Azevedo

Thuerey

et al. 2019

Computer Graphics Forum

295

267

View full text Add to dashboard Cite

This paper presents a novel generative model to synthesize fluid simulations from a set of reduced parameters. A convolutional neural network is trained on a collection of discrete, parameterizable fluid simulation velocity fields. Due to the capability of deep learning architectures to learn representative features of the data, our generative model is able to accurately approximate the training data set, while providing plausible interpolated in‐betweens. The proposed generative model is optimized for fluids by a novel loss function that guarantees divergence‐free velocity fields at all times. In addition, we demonstrate that we can handle complex parameterizations in reduced spaces, and advance simulations in time by integrating in the latent space with a second network. Our method models a wide variety of fluid behaviors, thus enabling applications such as fast construction of simulations, interpolation of fluids with different parameters, time re‐sampling, latent space simulations, and compression of fluid simulation data. Reconstructed velocity fields are generated up to 700× faster than re‐simulating the data with the underlying CPU solver, while achieving compression rates of up to 1300×.

show abstract

“…In the field of fluid simulation, Ladický et al employ the regression forest to approximate the acceleration of fluid particles in real‐time SPH. Yang et al utilized an artificial NN to avoid a time‐consuming projection step in the Eulerian method; a similar method is also proposed in the work of Tompson et al with convolutional NN. Furthermore, machine learning technology is successfully applied to patch‐based detail enhancement, superresolution smoke, deformation‐aware fluid, and a Lattice Boltzmann method (LBM) .…”

Section: Related Workmentioning

confidence: 99%

Data‐driven retrieval of spray details with random forest‐based distance

Chen

Zipeng

et al. 2019

Computer Animation & Virtual

View full text Add to dashboard Cite

Generating realistic spray details in liquid simulations remains computationally expensive. This paper proposes a data‐driven method to simulate high‐resolution sprays on low‐resolution grids by retrieving details with the most compatible details from a precomputed repository efficiently. We first employ a random forest‐based distance (RFD) to measure the similarity of liquid regions. In consideration of spatiotemporal relationships between one liquid region and its neighbors, we define a multinary label for RFD instead of the original binary one. Our improved RFD enables us to retrieve details that fit ground truth the best. To ensure temporal continuity of our result and to generate new details from existing ones, we formulate a series of forests with a training set from different time steps. Then, we synthesize results of each forest according to their distances. Finally, we put the synthesis result in correct positions to generate desired sprays motion. In our method, a state‐of‐the‐art cascade forest is employed for a higher accuracy. Several experiments with various grid resolutions validate our method both in visual effect and computational cost.

show abstract

Data‐driven projection method in fluid simulation

Cited by 104 publications

References 23 publications

Latent Space Physics: Towards Learning the Temporal Evolution of Fluid Flow

Latent Space Physics: Towards Learning the Temporal Evolution of Fluid Flow

Deep Fluids: A Generative Network for Parameterized Fluid Simulations

Data‐driven retrieval of spray details with random forest‐based distance

Contact Info

Product

Resources

About