Ruprecht-Karls-Universität Heidelberg
Uni Heidelberg > IWR > Newsroom
[english] | [deutsch]


The IWR is an interdisciplinary research center for Scientific Computing which builds bridges across disciplines. It promotes mathematical and computational methods in science, engineering and the humanitites. Currently the IWR comprises more than 50 research teams from various faculties. Around 600 scientists work together in interdisciplinary cooperation projects. In addition to educating the next generation of excellent scientists the IWR also focuses on advacing international research networks.


May 22, 2021

ERC Advanced Grant for Prof. Anna Wienhard

Congratulation to IWR Member Prof. Anna Wienhard on being awarded a highly endowed grant from the European Research Council (ERC). The ERC is making approx. two million euros available to Prof. Wienhard, who teaches and does research work at the Mathematical Institute of Heidelberg University, for her project on symmetries in mathematics.

The ERC is funding Prof. Wienhard’s research for the second time. In 2014 she received an ERC Consolidator Grant for her work on symmetries and deformation spaces of geometric structures. The present project “PosLieRep – Positivity in Lie Groups and Representation Varieties” builds on the previous one and focuses on Lie groups, which play a central role in many fields of mathematics and are an important tool in theoretical physics. Lie groups describe the symmetries of a space or a system. An important structure in Lie groups is total positivity, which was developed around 1930 in connection with vibrations of mechanical systems. It has many applications in discrete mathematics, in the theory of stochastic processes and in representation theory. Prof. Wienhard has discovered new positivity structures that generalise the total positivity in Lie groups, and the PosLieRep project is to study these new structures. “We hope to gain new insights from this, in particular on the theory of higher Teichmüller spaces. At the same time, this field of research opens up interesting new perspectives on further areas, for instance super-symmetrical field theories in physics,” the scientist explains.

Anna Wienhard studied theology and mathematics at the University of Bonn and obtained her doctorate in mathematics there in 2004. There followed various research stays in Switzerland and the United States, including at Princeton University, where she was an assistant professor from 2007 on. In 2012 Anna Wienhard accepted a professorship at Heidelberg University. At the Mathematical Institute she heads the working group on differential geometry and also the geometry and dynamics research station. Furthermore, she is co-spokesperson of the STRUCTURES Cluster of Excellence and a member of the Interdisciplinary Center for Scientific Computing. Since 2015, she has also been a group leader at the Heidelberg Institute for Theoretical Studies (HITS). Prof. Wienhard has received a number of awards and grants for her research projects.

[Full Press Release Heidelberg University][Website Prof. Wienhard]

April 7, 2021

Using AI to Diagnose Neurological Diseases Based on Motor Impairment

New Heidelberg approach: analysing movement patterns through machine learning

The way we move says a lot about the state of our brain. While normal motor behaviour points to a healthy brain function, deviations can indicate impairments owing to neurological diseases. The observation and evaluation of movement patterns is therefore part of basic research, and is likewise one of the most important instruments for non-invasive diagnostics in clinical applications. Under the leadership of computer scientist Prof. Dr Björn Ommer and in collaboration with researchers from Switzerland, a new computer-based approach in this context has been developed at Heidelberg University. As studies inter alia with human test persons have shown, this approach enables the fully automatic recognition of motor impairments and, through their analysis, provides information about the type of the underlying diseases with the aid of artificial intelligence.

For the computer-supported movement analysis, subjects usually have to be tagged with reflective markings or virtual markers have to be applied to the video material produced in the framework of the assessment. Both procedures are comparatively complicated. Furthermore, conspicuous movement behaviour has to be known in advance so that it can be further examined. “A real diagnostic tool should not only confirm motor disorders but be able to recognise them in the first place and classify them correctly,” explains Prof. Ommer, who heads the Computer Vision group at the Interdisciplinary Center for Scientific Computing at Heidelberg University.

Precisely that is made possible by the novel diagnostic method developed by his team, and known as “unsupervised behaviour analysis and magnification using deep learning” (uBAM). The underlying algorithm is based on machine learning using artificial neural networks and it recognises independently and fully automatically characteristic behaviour and pathological deviations, as the Heidelberg scientist explains. The algorithm determines what body part is affected and functions as a kind of magnifying glass for behavioural patterns by highlighting different types of deviation directly in the video and making them visible. As part of this, the relevant video material is compared with other healthy or likewise impaired subjects. Progress in treating motor disorders can also be documented and analysed in this way. According to Prof. Ommer, conclusions can also be drawn about the neuronal activity in the brain.

The basis for the uBAM interface is a so-called convolutional neural network, a type of neural network that is used for image recognition and image processing purposes especially. The scientists trained the network to identify similar movement behaviour in the case of different subjects, even in spite of great differences in their outward appearance. That is possible because the artificial intelligence can distinguish between posture and appearance. Besides the recognition and quantification of impairments, a detailed analysis of the symptoms is also important. “To study them in detail, we use a generative neural network,” says Prof. Ommer. “That way we can help neuroscientists and clinicians focus on subtle deviations in motor behaviour that are likely to be overlooked with the naked eye, and make them easily visible by magnifying the deviation. Then we can exactly demarcate the type of disease in the individual case.”

The research team has already been able to prove the effectiveness of this new approach with the aid of different animal models and studies with human patients. They tested, inter alia, the precision with which uBAM can differentiate between healthy and impaired motor activity. In their publication on the topic, the scientists report a very high retrieval rate both in mice and human patients. “In all, our study shows that, as compared to conventional methods, the approach based on artificial intelligence delivers more detailed results with significantly less effort,” Björn Ommer emphasises.

With respect to the application, the scientists hope that uBAM will be used both in basic biomedical research and in clinical diagnostics and beyond. Prof. Ommer: “The interface can be applied where traditional methods prove too complicated, tedious, or not efficient enough. Potentially it could lead to a better understanding of neuronal processes in the brain and the development of new therapeutic options.”

Besides the Heidelberg researchers working with Prof. Ommer, scientists from the University of Zurich and University Hospital Zurich, Balgrist University Hospital and the Neuroscience Center Zurich were also involved in developing the uBAM interface. Part of the funding for the study came from the German Research Foundation as well as the Branco Weiss Fellowship Society in Science and the Swiss National Foundation. The results were published in the journal “Nature Machine Intelligence”.

[Press Release Heidelberg University][Website Prof. Ommer][Link Paper]

26. März 2021

3D-Scannen von Inschriften einer mittelalterlichen Altarplatte in der Kirche St. Peter und Paul in Niederzell, UNESCO Welterbe Klosterinsel Reichenau

IWR-Mitglied Dr. Susanne Krömker und ihre Forschungsgruppe machen Geschichte mithilfe modernsters Technik sichtbar.

Im März 2021 wurde mit dem hochauflösenden 3D-Scanner des IWR (Streifenlichtscanner smartScan3D-HE, AICON/Hexagon) eine frühmittelalterliche Altartafel in der Kirche St. Peter und Paul in Niederzell, UNESCO-Welterbe Klosterinsel Reichenau aufgenommen. Diese Altartafel wurde erst 1976 bei Restaurierungsarbeiten entdeckt, da sie Jahrhunderte lang verkehrt herum auf einem gemauerten Altar Verwendung gefunden hatte. Seitdem war sie zwar mit der Inschriftenseite nach oben an gleicher Stelle bewahrt, jedoch unter einer massiven Eichenholzkonstruktion wieder verschwunden. Die rund 300 Inschriften mit den Namen von Mönchen (vergleichbar mit den sogenannten Verbrüderungsbüchern) oder von Pilgern auf dem Weg nach Rom können alle auf die Zeit um 800 bis 1000 n. Chr. datiert werden. Sie sind in den Rorschacher Sandstein eingeritzt oder mit dickflüssiger Tinte geschrieben. Derart mit Inschriften versehene Altartafeln sind im deutschen Raum einzigartig, finden sich ansonsten nur in einigen Kirchen Südfrankreichs und auch dort nicht mit einer so großen Anzahl erhaltener Namen. Mit Tinte geschriebene Namen sind nur auf der Platte in Niederzell gefunden worden.

Die Verarbeitung der Daten mit den am IWR entwickelten Methoden (u.a. mit der GigaMesh-Software von Hubert Mara) ist Teil der laufenden Forschung.

Die Ausrüstung wurde durch die Heidelberg Graduate School of Mathematical and Computational Methods for the Sciences (HGS MathComp) zur Verfügung gestellt.

[Webseite des Projekts][Homepage Dr. Krömker]

March 1, 2021

Understanding the Spatial and Temporal Dimensions of Landscape Dynamics

Heidelberg geoinformation scientists develop new computer-based method to analyse topographic changes

The Earth’s surface is subject to continual changes that dynamically shape natural landscapes. Global phenomena like climate change play a role, as do short-term, local events of natural or human origin. The 3D Geospatial Data Processing (3DGeo) research group of Heidelberg University has developed a new analysis method to help improve our understanding of processes shaping the Earth’s surface like those observed in coastal or high-mountain landscapes. Unlike conventional methods that usually compare two snapshots of the topography, the Heidelberg approach can determine – fully automatically and over long periods – when and where surface alterations occur and which type of associated changes they represent.

The method, known as spatiotemporal segmentation, was developed under the guidance of Prof. Dr Bernhard Höfle, whose 3DGeo group is based at the Institute of Geography and the Interdisciplinary Center for Scientific Computing (IWR) of Heidelberg University. “By observing entire surface histories, our new computer-based method allows for more flexible approaches. Unlike with previous methods, we no longer have to specify which individual change processes we want to detect or the points in time the analysis should include,” the geoinformation scientist states. “Instead, areas and entire time periods during which similar changes occur are identified fully automatically. The huge three-dimensional datasets from the automatic laser measurements in the landscape thereby reveal various types of changes that the direct comparison of only two measurement points does not.”

Among other techniques, Prof. Höfle’s team uses terrestrial laser scanning (TLS) to measure mountain and coastal landscapes. It generates three-dimensional models of a landscape represented as billions of measurement points in so-called 3D point clouds. “Measurement systems are installed on site and capture the terrain in short, regular intervals over several months, thus generating three-dimensional time series,” explains Katharina Anders, a PhD student in Bernhard Höfle’s research group and at the IWR of Heidelberg University. These 3D time series are special because they contain both the temporal and spatial – ergo 4D – properties of surface changes, which can then be reviewed as in a time-lapse video.

“Spatiotemporal segmentation allows us to differentiate in detail between various phenomena that conventional methods detect as a single event or sometimes not at all,” states Katharina Anders.

The Heidelberg geoinformation scientists applied their method to a 3D time series of a stretch of coast in the Netherlands, which was acquired hourly over five months by scientists of the Delft University of Technology. The data analysis of the entire observation period revealed more than 2,000 changes representing temporary accumulation or erosion of sand that occurred in different locations at varying magnitudes and across various time periods. In this case, the dynamic transport of sand recorded by the measurement system was caused by complex interactions of wind, waves, and human influence. As a result, several truckloads of sand were transported on average in an area of 100 square metres over a period of four weeks, without influence from major storm events.

Findings of such analyses provide the basis for further studies of specific phenomena or underlying processes. At the same time, the information obtained on the dynamic evolution of surfaces opens up new possibilities for parameterisation and hence adaptation of computer-based environmental models. “The method we developed therefore makes an overall contribution to improving our geographic understanding of natural landscape dynamics,” adds Katharina Anders.

The results of the joint study with Delft University of Technology were published in the “ISPRS Journal of Photogrammetry and Remote Sensing”.

[Press Release Heidelberg University][Website Prof. Höfle][Link Publication]

February 4, 2021

Training AI Systems for Use in Dangerous Situations

Heidelberg researchers join in trilateral collaborative project.

Systems based on artificial intelligence must also function reliably in infrequent and dangerous situations. Researchers at Heidelberg University along with scientists from France and Japan are studying how to “train” AI systems for such circumstances. Prof. Dr Carsten Rother of the Interdisciplinary Center for Scientific Computing (IWR) is heading up the Heidelberg arm of the collaboration, which is being funded with approximately 300,000 euros by the German Research Foundation for a period of three years.

“AI-based systems will find their way into many areas of our life in the near future, coexisting with humans in dynamically changing environments,” states Prof. Rother. “In spite of considerable progress in the field of machine learning, these types of systems do not always perform reliably in all scenarios, especially in dangerous situations that don’t arise every day.” Given this, the researchers are planning to develop dedicated software and hardware in order to generate new data for machine learning based on images and video footage. This will allow researchers to specifically generate rare scenarios to test and improve their artificial intelligence applications.

Also participating in the joint project, entitled “Understanding and Creating Dynamic 3D Worlds towards Safer AI”, are the École des Ponts ParisTech and Japan’s Kyoto University. It is one of nine trilateral projects in the field of artificial intelligence being funded in the amount of approximately seven million euros by the Agence Nationale de la Recherche and the Japan Science and Technology Agency along with the German Research Foundation. Scientists from Germany, France and Japan are participating in each of these projects.

[Press Release Heidelberg University][Website Prof. Rother]
Last Update: 05.05.2021 - 11:27


Jan Keese
Communications & Organization



Akademische Mittagspause 2016 "Sprechen Sie Mathematik?"

HGS MathComp Alumna Emőke-Ágnes Horvát on DW-TV.

Lange Nacht der Robotik 2013 in Heidelberg.