Semantic Modelling and Acquisition

Acquisition, understanding and planning of real-world environments made easy.

Research Priorities

Our group of highly motivated researchers builds software prototypes to solve real-world problems for a number of industry partners, and also performs applied and basic research resulting in publications at international conferences and journals.
The group's overarching goal is to make it easy to capture, represent, visualize and understand vast amounts of data, which are readily generated at almost arbitrary detail with today's technologies.
In order to cope with all these data and computation at interactive rates to accelerate decision-making processes, we combine methods from geometric modeling, visual analytics, simulation, human-computer interaction, virtual reality and rendering.
Finally, we also successfully engage in basic research of visualization pipelines as well as smart modeling systems with a focus on models of the real world.

Competences

  • High-performance visualization pipelines
  • Interactive processing, editing and visualization of very large datasets
  • Decision-making tools in multi-modal data scenarios
  • Visual analytics for spatial data enriched with heterogeneous metadata

Key Publications

Sorger, J, Ortner, T, Luksch, C, Schwärzler, M, Gröller, E, and Piringer, H (2015).
LiteVis: Integrated Visualization for Simulation-Based Decision Support in Lighting Design
IEEE Transactions on Visualization and Computer Graphics / Proceedings of IEEE VAST 2015 .

Luksch, C, Tobler, RF, Habel, R, Schwärzler, M, and Wimmer, M (2013).
Fast Light-Map Computation with Virtual Polygon Lights
in Proceedings of ACM Symposium on Interactive 3D Graphics and Games 2013 (i3D 2013):pp. 87-94.

Haaser, G, Steinlechner, H, Maierhofer, S, and Tobler, RF (2015).
An Incremental Rendering VM
In: Proccedings of HPG 2015

Arikan, M, Schwärzler, M, Flöry, S, Wimmer, M, and Maierhofer, S (2013).
O-Snap: Optimization-Based Snapping for Modeling Architecture
ACM Transactions on Graphics (TOG), 32(1):6:1-6:15.

Wörister, M, Steinlechner, H, Maierhofer, S, and Tobler, RF (2013).
Lazy Incremental Computation for Efficient Scene Graph Rendering
in Proceedings of High-Performance Graphics (HPG 2013):53-62.

Visual Computing Software-Plattform Aardvark

Our projects can draw from a huge treasury of existing software components. We don’t keep re-inventing the wheel: Aardvark combines valuable results from many VRVis projects into a powerful platform of programs and libraries for a variety of applications in the field of visual computing, and can be applied to a whole range of areas. This includes intelligent geometrical reconstruction and modeling tools, physically accurate lighting design in real time, as well as interactive visualizations of infrastructure projects or planetary scans from NASA/ESA missions.

A remarkable strength of Aardvark is the flexibility and dynamism in terms of data: easy-to-use high-level interfaces, self-optimizing visualization speedup techniques, incremental processing of dynamic real-time data, and out-of-core strategies allow the handling of massive amounts of industrial-scale data and information - and it is not limited to typical game-level scale.

Finally, a growing number of Aardvark libraries is available as open source on Github and being developed in the open (https://github.com/vrvis).

References

Zumtobel Lighting GmbH
rmData GmbH
Witsch Visuals GmbH

Universität Innsbruck, Institut für Konstruktion und Materialwissenschaften
Technische Universität Ilmenau
HAW Hamburg
Kunsthistorisches Museum Wien
IST Austria, Computergrafik
Österreichische Akademie der Wissenschaften, Institut für Schallforschung

Projects