Simulation and Digital Twin

Prof. Barbara Solenthaler from ETH Zurich, Switzerland, works as a Hans Fischer Fellow with the TUM host Prof. Nils Thürey in the Focus Group Simulation and Digital Twin. The group pursues research on capturing fluids and acquiring a full flow description. Existing capture approaches typically require complex hardware setups and calibration, and hence there is a strong demand for capturing volumetric flow phenomena with more simple systems. At the same time, there is a huge lack of reliable flow databases that can be used for ground truth comparisons or for developing data-driven techniques. Our TUM-IAS project fills this gap: We extend the existing video-based flow capture stage at Thürey’s group at TUM to a multi-camera framework and compute accurate tomography reconstructions. We will capture a large number of high-resolution datasets and create the first large scale and publicly available flow database. The real-world captured data will be connected to physics simulations by deep learning methods that allow a high-quality mapping between the different data sources. Additionally, we target solving inverse problems in the simulation context and address challenges related to transfers between data domains.

Prof. Solenthaler holds a TUM-IAS Hans Fischer Fellowship funded by Siemens AG.

TUM-IAS funded doctoral student:

Erik Franz,

Selected Publications by the Focus Group


  • Franz, Erik; Solenthaler, Barbara; Thuerey, Nils: Learning to Estimate Single-View Volumetric Flow Motions without 3D Supervision. , 2023 mehr…


  • Franz, Erik; Solenthaler, Barbara; Thuerey, Nils: Global Transport for Fluid Reconstruction with Learned Self-Supervision. 2021 mehr…


  • Wiewel, S.; Kim, B.; Azevedo, V. C.; Solenthaler, B.; Thuerey, N.: Latent Space Subdivision: Stable and Controllable Time Predictions for Fluid Flow. Computer Graphics Forum 39 (8), 2020, 15-25 mehr…


  • Kim, Byungsoo; Azevedo, Vinicius C.; Thuerey, Nils; Kim, Theodore; Gross, Markus; Solenthaler, Barbara: Deep Fluids: A Generative Network for Parameterized Fluid Simulations. Computer Graphics Forum 38 (2), 2019, 59-70 mehr…