All images and also data acquired by sensors can be converted into a digital representation. Digital twins are generated from real-world data, which serve as the basis for several digital applications. In our research work we focus on the fields of remote sensing and semantic reconstruction as well as on solutions for inclusive digitization and the medical sector.
Remote sensing enables us to obtain information about places or objects that would otherwise not be accessible and analyzable by recording spatial data. These can be rock walls or facades photographed by drone flight, as well as entire continents via satellite images or distant planets that can be reconstructed from measured data. We create solutions to adapt remote sensing methods for different industries. Our research results range from interactive visualization of the surface of Mars for planetary research/space travel to the generation of BIM models from existing scans of existing structures. The innovative bridging between data acquisition and semantic modelling is important to us.
In the field of semantic reconstruction, we are working, among other things, on finding efficient ways to automatically segment and classify buildings or other objects on images using computer vision algorithms and machine learning, in order to then "intelligently" reconstruct them as a three-dimensional model. This means, for example, that the 3D model of a house itself "recognizes" where its walls, doors, windows and so on are. Here, we optimize semi-automatic approaches that are controlled by the user on the one hand, and work on the other hand on the development of fully automatic methods.
In order to make the museums of the future barrier-free, we have been conducting intensive research for many years in the field of inclusive digitization to find IT-based solutions that can make the fine arts "visible" to all target groups. To this end, we have developed, among other things, special visualization software that creates a digital 2.5D model from 3D scans or photographs of museum objects. Using an innovative milling process, this model can be transformed into a tactile relief that makes heights, depths, surface structures, etc. haptically perceptible - thus equipping museums for the 21st century.
For the efficient further use and analysis of data obtained by imaging methods in the medical sector, we are researching innovative ways of automatically segmenting specific image content such as vertebrae, tumors or organs using machine learning and specially developed algorithms. This is an important supporting measure in the diagnostic process and also serves as a basis for reconstructions of volume data, for example.
A snap algorithm based on optimization automatically creates polygonal models from unstructured point clouds. These reconstructions can then be refined by the user with simple 2D strokes and are suitable for use in interactive 3D applications such as games or virtual environments.
The tool 3DWorx, developed together with our industry partner rmDATA, not only processes point clouds from laser scans or photogrammetry of buildings to models, which is of great importance for BIM projects, but also maps the entire reconstruction chain from scan to finished product. In addition, the derivation of relevant geometries is easily possible, which facilitates the creation of submission plans etc.
Together with our industrial partner dibit Messtechnik GmbH and other research partners, we have been involved for years in the development of software solutions for the visual analysis of extremely high-resolution 3D tunnel models. Based on these reconstructions, for example, an exact geotechnical monitoring from the desk is possible, whereby lengthy and expensive tunnel inspections on site can be minimized.
The HILITE software, which we have developed over many years together with our industrial partner ZUMTOBEL, enables the simulation of planned lighting systems directly from IFC files, combined with various data on geometry, material or light sources. This is done in physically exact and high-quality real-time visualizations, which can also be used to run through optimization versions. Originally developed together with our project partner ÖBB specifically for new BIM planning methods for railway stations, the lighting simulation software HILITE is now used in a number of other scenarios.
In the PROVEX project, JOANNEUM Research and VRVis are developing a provenance-aware workflow management tool for heterogeneous data for use in ESA's HERA mission.
No blind spots on train roofs thanks to depth cameras.
PanCam-3D focuses on the further development of interactive 3D visualizations for the ExoMars 2022 mission.
WIBSTAC addresses the usage of wide baseline stereo 3D reconstruction for medium- and long-range mapping of the Martian surface, based on imagery from panoramic rover camera instruments.
With the help of 3D printing (additive manufacturing), spare parts for defective trains can be produced more easily and in a sustainable way as well as faster and cheaper - a great potential for the climate-friendly future of train transport companies.
Together with Rhomberg Bau GmbH and convex ZT GmbH, VRVis is developing a concept for the use of Boston Dynamic's robot dog "Spot" for autonomous, immersive construction site documentation.
The research goal of AMASE is to create a suite of tools and methods to ingest, process, visualize, and manipulate heterogeneous, large-scale geospatial data. This data is the constantly updated representation of the real world in the form of an evolving digital twin.
The main goal of this project is to enable a reliable decision support for large-scale infrastructure projects by providing solutions for a collaborative visual analysis of digital twins.
The main objective of the strategic project ARCS is the design of software architectures that enable interactive visualization systems to ingest large volumes and velocities of geospatial and associated non-geometric data.
The applied research project Lightbox 2.0 focuses on the development of a photogrammetric 3D scanner for automatic and deep learning-based modeling of all kinds of keys.
The "Mars-DL" project is investigating how a deep learning system can support the research work of planetary scientists through object and pattern recognition. For this project VRVis has extended the functionality of PRo3D to automatically render shatter cone training images.
Support for planetary research: Visual analysis of reconstructions of the Mars surface and view planning for rover camera instruments.
Visual Analysis of Asteroid Deflection.
Virtual exploration and geological analysis of reconstructed Mars surfaces and rock outcrops.
In this project tools and methods for handling, administration, manipulation and evaluation of several different data sources for measurements and lighting design are developed.
MINERVA is an integrated framework for planetary scientists allowing members of different instrument teams to cooperate synergistically in virtual workspaces by sharing observations, analyses and annotations of heterogonous mission data.
Barrier-free access to art and museums for blind and visually impaired people through 3D technology.
Strategic Research in Scalable, Semantic Rendering.
Planetary Robotics Vision Data Exploitation.
Investigation of techniques enabling a seamless analysis of data from multi-run simulations on multiple degrees of detail.
Algorithms to improve the visual analysis of surface reconstructions.
Decision support systems and 3D viewing technologies for tunnel construction.