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

Team

Behind our smart solutions and software are smart and driven people.
Do you want to meet the team?

Publications

Harald Steinlechner, Bernhard Rainer, Michael Schwärzler, Georg Haaser, Attila Szabo, Stefan Maierhofer, Michael Wimmer (2019)
Adaptive Point-cloud Segmentation for Assisted Interactions
Symposium on Interactive 3D Graphics and Games (I3D) 2019

A. Walch, C. Luksch, A. Szabo, H. Steinlechner, G. Haaser, M. Schwärzler, S. Maierhofer (2018).
Lens flare prediction based on measurements with real-time visualization
The Visual Computer

K.Krösl, A.Felnhofer, J.X.Kafka, L.Schuster, A.Rinnerthaler, M.Wimmer, O.Kothgassner (2018).
The Virtual Schoolyard: Attention Training in Virtual Reality for Children with Attentional Disorders
Siggraph 2018 (poster, semi-finalist)

K. Krösl, D. Bauer, M. Schwärzler, H. Fuchs, G. Suter M. Wimmer (2018).
A VR-based User Study on the Effects of Vision Impairments on Recognition Distances of Escape-Route Signs in Buildings
The Visual Computer

A. Walch, K. Krösl, C. Luksch, D. Pichler, T. Pipp, M. Schwärzler (2018).
An Automated Verification Workflow for Planned Lighting Setups using BIM
REAL CORP 2018, Proceedings

K. Krösl, C. Luksch, M. Schwärzler, M. Wimmer (2017).
LiteMaker: Interactive Luminaire Development using Progressive Photon Tracing and Multi-Resolution Upsampling
in Proceedings of Vision, Modeling & Visualization

Visual Computing Software-Platform Aardvark & Licensing

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/aardvark-platform/aardvark.docs/wiki/components). We offer the Aardvark software components under various licenses which correspond to different needs: industrial development, industrial and academic research. Learn more about the licensing options here.