User: Guest  Login
Title:

Learning Similarity Metrics for Volumetric Simulations with Multiscale CNNs

Document type:
Forschungsdaten
Publication date:
29.03.2023
Responsible:
Kohl, Georg
Authors:
Kohl, Georg; Chen, Li-Wei; Thuerey, Nils
Author affiliation:
TUM
Publisher:
TUM
Identifier:
doi:10.14459/2023mp1703144
End date of data production:
01.02.2023
Subject area:
DAT Datenverarbeitung, Informatik; PHY Physik
Resource type:
Simulationen / simulations
Data type:
Datenbanken / data bases
Other data type:
Compressed numpy arrays of dense volumetric fields
Description:
This archive contains volumetric data from different PDE simulations. It can be used to train and evaluate the similarity assessment of different metrics.
Method of data assessment:
This data set also contains a selection of data from the Johns Hopkins Turbulence Database
( http://turbulence.pha.jhu.edu) as detailed in our source code.
Links:

Project: https://ge.in.tum.de/publications/2022-volsim-kohl
Publication: https://arxiv.org/abs/2202.04109
Source Code: https://github.com/tum-pbs/VOLSIM

Key words:
Deep Learning; Similarity Assessment; Fluid Simulation
Technical remarks:
View and download (89 GB total, 16 Files)
The data server also offers downloads with FTP
The data server also offers downloads with rsync (password m1703144):
rsync rsync://m1703144@dataserv.ub.tum.de/m1703144/
Language:
en
Rights:
by, http://creativecommons.org/licenses/by/4.0
Horizon 2020:
ERC 2019 CoG 863850 (SpaTe)
 BibTeX