With our expertise, we are currently contributing to a number of research projects. While fundamental research is supported by public funding agencies, applied research is funded by industry. Our current projects are listed here, completed projects can be found at the end of this page.
Chair of Visual Computing
SMOOTH - Spatial and Temporal Filtering of Depth Data for Telepresence
In the past years, there has been a growing interest in telepresence communication systems, which create the impression of being present at a place different from the true location. A major challenge in this area is to process the acquired imagery at the sender site into a high-quality 3D representation of the scene in real time. High quality approaches usually require intensive offline processing. Then again, methods that work in real time produce 3D representations at a low grade. In recent systems this is caused by the application of low-cost depth sensors, such as Microsoft Kinect, which deliver 3D representations in real time but still exhibit considerable amounts of disturbing artifacts. The flickering nature of artifacts is often not considered. To understand their strong temporal component, we will start with experimentally developing a statistical model for the distortions of common depth sensors. In contrast to existing work in this field, we will also consider the temporal aspects of the matter. Guided by the results of analyzing the gained data, we will develop a new real-time spatio-temporal filter to simultaneously stabilize distorted depth data in the spatial and temporal domains. Therefore, we suggest a composition of a novel depth outlier detection method, motion estimation for depth cameras and 3D-filtering. more...
ARGuide - Influence of Gravity on Exocentric and Egocentric Target Cueing Using Augmented Reality
2016-2018, Bundesministerium für Wirtschaft und Technologie (BMWi)
During payload operations astronauts are guided by sequential directives displayed on a laptop computer using an exocentric presentation scheme for task guiding. Such an approach forces the astronaut to constant changes of focus that can cause loss of concentration and attention, as well as can be the primary reason for sequence errors resulting in a faulty task termination. To ease astronauts’ work and ensure successful task performance new interface technologies are required. By bridging the gap between the physical reality and digital information, Augmented Reality (AR) keeps the focus on the task to fulfill and offers user- centered operations by an egocentric display. Beside the display, AR interface can differ in providing the visual information for localizing users’ attention. While egocentric visualizations maintain the principal characteristic required for AR interfaces by 3D registered information, exocentric visualizations are presented as head-up display information and approved methods to navigate the user towards off-screen objects. Using a visuomotor task (visual search, operation task) we will investigate the influence of altered gravity on human performance and workload by comparing exocentric with egocentric displays and presentation schemes in a within-subject user study. To differentiate visuomotor deficits we will use common performance metrics. more...
Chair of Computer Graphics
UniVA - A Unified Interface for Visual Analytics
2018-2020, DFG Project, in cooperation with Prof. B. Urban, Fraunhofer IGD Rostock
Existing Visual Analytics (VA) tools are often special-purpose solutions that are tailored to a particular analysis scenario. Generic VA frameworks are rare, as they are complex to maintain and use. And eventually these frameworks are still lacking as it is impossible to foresee and meet any conceivable data analysis need. Hence, the idea of this project is to combine and coordinate existing VA tools, rather than developing a new “super-application”. For this purpose, we will provide a unified user interface that allows access to data and functionality of multiple VA tools to be used within a given workflow. The Computer Graphics workgroup will address this objective on the visual interface level by investigating view ensembles - i.e., the arrangement and interactive linking of visual output generated by different VA tools.
GEMS 2.0 - Visual Editing and Comparison of Multivariate Graphs Using Multiple Interactive Displays
2017-2020, DFG Project (Continuation of GEMS), in cooperation with Prof. R. Dachselt, TU Dresden
Multivariate Graphs are used in different application areas, for example for biological and social networks. This kind of graph associates data attributes with nodes and edges. Understanding such graphs comprehensively is a significant challenge for the people working with them. GEMS 2.0 focuses on developing new concepts, methods and tools for comparing and editing multiple multivariate graphs with novel multimodal interactions in multi-display environments. The main emphasis’ of the Rostock workgroup are: (1) Investigation of new visual representations to communicate the characteristics of the graphs structure and the associated data attributes. (2) Research of new concepts to edit nodes, edges and data attributes as well as a visual communication of the changes. (3) User support for the comparison and editing of multiple graphs.
TOPOs: Therapy prediction through OCT and und patients demographic data in the field of ophthalmology
2017-2020, BMBF project, in coop. with Prof. Ritter, (University of applied Science Mittweida), Prof. Dr. Stahl (Medical Center – University of Freiburg), Dr. med. Philipp Daumke (AVERBIS GmbH)
Optical Coherence Tomography (OCT) has been a suitable diagnostic tool for ophthalmologists since many years. Pictures, taken from the latter part of the eye, the retina, and vertical depth scans of the macula allow reliable diagnostics of a variety of eye diseases as for example age-related macular degeneration, retinal vein occlusion or diabetic retinopathy. Up to today there are successful therapy methods, first of all the injection of anti - vascular endothelial growth factor (anti-VEGF) drugs. Nevertheless it seems not possible to this date, to reliably predict the correct amount of injections necessary as well as the time needed until the injection rate can be decreased or halted. In order to find remedy for this issue, this project unites experts of ophthalmology and visual analytics. The goal is, to explore the heterogenic multidimensional data of the OCT examinations in combination with the patients historical and personal characteristics. In adjunction with a specialized visualization tool for ophthalmologists, the possible causes and/or influencing factors for the therapy of retinal diseases are expected to crystalize from the data.
VisSect: Visual Segmentation and Labeling of Multivariate Time Series
2016-2019, DFG project, in coop. with TU Darmstadt and TU Wien
Segmentation and labeling of multivariate time series is a highly relevant task in many domains, for instance, in the analysis of human motion data or electrocardiograph data. The goal in such analyses is to find meaningful segments in time series and compare them with similar segments. This research project aims to develop new approaches for visually and interactively analyzing multivariate time series in order to support the generation of hypotheses and the drawing of conclusions. In this regard, the overarching objective is the combination of three relevant steps in this process: (1) algorithm selection for segmentation and labeling, (2) parametrization, and (3) visualization and exploration of uncertainties about the results.
ViES: Visual-interactive Exploration for Individualized Selection of Relevant Data
2016-2018, DFG project, in coop. with Prof. Jünemann (Heidelberg Engineering GmbH) and Prof. Stachs (University of Rostock)
This project aims at supporting the diagnosis of glaucoma induced retinal damage by interactive visual means. The industrial application partner, Heidelberg Engineering GmbH, provides high-resolution tomographic data sets that need to be reduced for their assessment. Currently, only a simplified automated data reduction is applied, which can lead to misinterpretations, particularly in case of small retinal changes. These automated computations will be extended with novel visual interactive methods. This way, smaller individualized data sets are generated that are easily manageable by physicians and allow for comprehensive analyses of retinal sub structures. For this project, we can take advantage of previously developed approaches in the DFG program "Scalable Visual Analytics".
Applications of computer graphics methods, theory and algorithms for multi-functional cockpit displays
2007-2019, industry project, Diehl Aerospace
Today's graphics hardware makes it possible to display high-quality graphics in real-time. Yet, always a compromise has to be found balancing computation time, use of resources, and obtained image quality. Especially in time-critical systems with limited resources rendering images with a guaranteed sub-pixel accuracy is a challenge. In this project we address this challenge with new solutions related to the rendering of large terrains taking into account the requirements of cockpit displays. An increasingly important aspect is the integration of additional data about current weather conditions or flight information into these displays.