Research projects

The research is pursued in close collaboration with our company partners and with TU Wien. We focus on simulation data and streaming data with all related challenges. Streaming data comes from complex simulations such as large-scale CFD, real-time monitoring, or simulation of sensors in a vehicle, for example, or from the acquisition of volumetric imaging data as in ultrasound in the medical domain. In addition, we aim at improving visualization techniques in general and at enhancing interactive visualization with different forms of domain-specific user guidance. When it comes to visualization improvements, the main goal is to research novel visualization methods for fast sensor simulation, for modern, meshless simulation results, for medical imaging in case of ultrasound, and in using deep learning and other AI methods in order to plausible predict simulation or measurement results and to use this prediction in visualization.

Our objective is also to develop novel progressive and quantitative visual analysis methods. Due to the increased complexity and size of datasets, it is often not convenient to wait for various computations to complete, to inspecting the results then, and to restart computation if necessary. Progressive visual analytics enables an analyst to inspect partial results of an algorithm as they become available and interact with the algorithm to prioritize subspaces of interest. As such, a prioritization is heavily domain- and task-dependent, and our objective is to research interactive methods, which support domain experts in progressive analytics workflows.

We further pursue our research on how to make the whole visual analysis process more quantitative. Novel methods that help the users to quantify analysis results are urgently needed, as they are the best way to increase comparison and communication of analysis results. Finally, based on our successful previous research and cooperation with the area "Immersive Analytics", we continue to research the use of non-conventional interfaces for analysis and presentation of simulation results.

The main objectives can be summarized as:

• Research of novel approaches that support in-situ visualization and analysis of streaming data originating from simulations, volumetric imaging data, and various sensors.

• Novel visualization methods for modern data sources such as meshless simulation or modern imaging devices.

• Research how deep learning-based user modeling can empower visualization and interaction.

• Extend current state of the art in quantitative and progressive visual analytics.

• Deploy non-conventional interfaces and XR techniques for analysis and presentation of simulation results.