Ullrich Köthe

[ Scientific Interests | VIGRA | CV | Publications | Teaching ]

Prof. Dr. Ullrich Köthe

Group Leader in the Visual Learning Lab Heidelberg
Mathematikon B (Berliner Str. 43), level 3, office A119

Email: ullrich.koethe (at) iwr.uni-heidelberg.de
Phone: +49 6221 54 14834
Fax: +49-6221-54 5276
Interdisciplinary Center for Scientific Computing (IWR)
University of Heidelberg
Im Neuenheimer Feld 205
69120 Heidelberg

Scientific Interests

I’m heading the subgroup on “Explainable Machine Learning”. Explainable learning shall open-up the blackbox of successful machine learning algorithms, in particular neural networks, to provide insight rather than mere numbers. To this end, we are designing powerful new algorithms on the basis of invertible neural networks and apply them to medicine, image analysis, and the natural and life sciences.

In addition, I’m interested in generic software bringing state-of-the-art algorithms to the end user and maintain the VIGRA image analysis library.


By individual arrangement

  • Master and bachelor theses in the field of machine learning and image analysis (Informatik, Scientific Computing, Physics)
  • Practicals, creditable for e.g. BSc Informatik (IFP), MSc Informatik (IFM), Physics (WPProj)

Summer Term 2022 (planned)

Winter Term 2021/22

Previous semesters

Selected Publications

Please refer to my profile at Google Scholar and the DBLP Citation Database for a more complete list. My pre-2010 publications can also be found here.


  • U. Köthe: “Reliable Low-Level Image Analysis”
    Habilitation Thesis, Department Informatik, University of Hamburg, 318 pages, Hamburg 2008
    Abstract | PDF (10 MB) 

  • U. Köthe: “Generische Programmierung für die Bildverarbeitung”
    PhD Thesis, Fachbereich Informatik, Universität Hamburg, 274 pages, Hamburg 2000, ISBN: 3-8311-0239-2. (in German)
    Abstract | PDF (12.5 MB) 

Recent and popular papers:

  • S. Radev, F. Graw, S. Chen, N. Mutters, V. Eichel, T. Bärnighausen, U. Köthe (2021). “OutbreakFlow: Model-based Bayesian inference of disease outbreak dynamics with invertible neural networks and its application to the COVID-19 pandemics in Germany”, PLOS Computational Biology, arXiv:2010.00300. [link], [arxiv], [pdf]
  • S. Radev, M. D’Alessandro, U. Mertens, A. Voss, U. Köthe, P. Bürkner (2021). “Amortized Bayesian model comparison with evidential deep learning”, IEEE Trans. Neural Networks and Learning Systems, arXiv:2004.10629. [link], [arxiv], [pdf]
  • P. Sorrenson, C. Rother, U. Köthe: “Disentanglement by Nonlinear ICA with General Incompressible-flow Networks (GIN)”, Intl. Conf. Learning Representations, 2020.
    Abstract | PDF
  • R. Mackowiak, L. Ardizzone, U. Köthe, C. Rother (2021). “Generative Classifiers as a Basis for Trustworthy Image Classification”, CVPR 2021 (oral presentation), arXiv:2007.15036 [arxiv], [pdf]
  • S. Radev, U. Mertens, A. Voss, L. Ardizzone, U. Köthe (2020). “BayesFlow: Learning complex stochastic models with invertible neural networks”, IEEE Trans. Neural Networks and Learning Systems, doi:10.1109/TNNLS.2020.3042395, arXiv:2003.06281. [link], [arxiv], [pdf]
  • S. Wolf, A. Bailoni, C. Pape, N. Rahaman, A. Kreshuk, U. Köthe, F.A. Hamprecht: “The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning”. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020.
    Link | PDF
  • L. Ardizzone, C. Lüth, J. Kruse, C. Rother, U. Köthe, “Guided Image Generation with Conditional Invertible Neural Networks”, arXiv:1907.02392, 2019.
    Abstract | PDF
  • S. Berg, D. Kutra, …, U. Köthe, F.A. Hamprecht, A. Kreshuk: “ilastik: interactive machine learning for (bio)image analysis”, Nature Methods, vol. 16, pp. 1226–1232, 2019.
  • L. Ardizzone, J. Kruse, S. Wirkert, D. Rahner, E.W. Pellegrini, R.S. Klessen, L. Maier-Hein, C. Rother, U. Köthe:
    “Analyzing Inverse Problems with Invertible Neural Networks”
    arXiv:1808.04730, Intl. Conf. Learning Representations, 2019.
    Abstract | PDF 

  • Stefan T. Radev, Ulf K. Mertens, Andreas Voss, Ullrich Köthe:
    “Towards end‐to‐end likelihood‐free inference with convolutional neural networks”
    British Journal of Mathematical and Statistical Psychology. doi: 10.1111/bmsp.12159, 2019.
    Abstract | PDF 

  • S. Wolf, C. Pape, A. Bailoni, N. Rahaman, A. Kreshuk, U. Köthe, F.A. Hamprecht:
    “The Mutex Watershed: Efficient, Parameter-Free Image Partitioning”
    in: Europ. Conf. Computer Vision (ECCV’18), pp. 546-562 , 2018.
    Abstract | PDF 

  • S. Wolf, L. Schott, U. Köthe, F.A. Hamprecht:
    “Learned Watershed: End-to-End Learning of Seeded Segmentation”
    in: Intl. Conf. Computer Vision (ICCV’17), pp. 2030-2038, 2017.
    Abstract | PDF 

  • C. Sommer, C. Straehle, U. Köthe, F.A. Hamprecht:
    “ilastik: Interactive learning and segmentation toolkit”
    In: IEEE International Symposium on Biomedical Imaging (ISBI), pp. 230-233, 2011.
    Abstract | PDF 

  • U. Köthe: “Edge and Junction Detection with an Improved Structure Tensor”
    in: B. Michaelis, G. Krell (Eds.): Pattern Recognition, Proc. of 25th DAGM Symposium, Magdeburg 2003, Springer LNCS 2781, pp. 25-32, 2003.
    Abstract | PDFAwarded the main prize of the German Pattern Recognition Society (DAGM) 2003 

  • U. Köthe: “Integrated Edge and Junction Detection with the Boundary Tensor”
    in: ICCV ‘03, Proc. of 9th Intl. Conf. on Computer Vision, Nice 2003, vol. 1, pp. 424-431, 2003.
    Abstract | PDF 

  • B. Andres, U. Köthe, M. Helmstaedter, W. Denk, F.A. Hamprecht:
    “Segmentation of SBFSEM Volume Data of Neural Tissue by Hierarchical Classification”
    in: G. Rigoll (Ed.): Pattern Recognition, Proc. DAGM 2008, Springer LNCS 5096 , pp. 142-152, 2008.
    Abstract | BibTeX | PDFReceived a Best Paper Award from the German Association for Pattern Recognition (DAGM) 

  • B. Andres, T. Kröger, K. Briggmann, W. Denk, N. Norogod, G. Knott, U. Köthe, F.A. Hamprecht:
    “Globally Optimal Closed-Surface Segmentation for Connectomics”
    in: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (Eds.) : 12th Eur. Conf. Computer Vision (ECCV 2012) part III, Springer LNCS 7574, pp. 778-791, 2012.
    Abstract | BibTeX | PDF 

  • T. Beier, B. Andres, U. Köthe, F.A. Hamprecht:
    “An Efficient Fusion Move Algorithm for the Minimum Cost Lifted Multicut Problem”
    in: Leibe, B., Matas, J., Sebe, N., Welling, M. (Eds.) : 14th Eur. Conf. Computer Vision (ECCV 2016), 2016.
    Abstract | PDF 

  • U. Köthe: “Reusable Software in Computer Vision”
    in: B. Jähne, H. Haussecker, P. Geissler (Eds.): Handbook of Computer Vision and Applications, Volume 3: Systems and Applications, pp. 103-132, San Diego: Academic Press, 1999.
  • U. Köthe, M. Felsberg:
    “Riesz-Transforms Versus Derivatives: “On the Relationship Between the Boundary Tensor and the Energy Tensor”
    in: R. Kimmel, N. Sochen, J. Weickert (Eds.): Scale Space and PDE Methods in Computer Vision, Springer LNCS 3459, pp. 179-191, 2005.
    Abstract | PDF 

  • A. Kreshuk, U. Köthe, E. Pax, D. Bock, F.A. Hamprecht:
    “Automated Detection of Synapses in Serial Section Transmission Electron Microscopy Image Stacks”
    PLoS ONE 9(2): e87351, 2014.
    Abstract | BibTeX | PDF 

  • B. Kausler, M. Schiegg, B. Andres, M. Lindner, U. Köthe, H. Leitte, J. Wittbrodt, L. Hufnagel, F.A. Hamprecht:
    “A discrete chain graph model for 3D+t cell tracking with high misdetection robustness”
    in: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (Eds.) : 12th Eur. Conf. Computer Vision (ECCV 2012) part III, Springer LNCS 7574, pp. 144-157, 2012.
    Abstract | BibTeX | PDF 

  • M. Hanselmann, U. Köthe, M. Kirchner, B.Y. Renard, E.R. Amstalden, K. Glunde, R.M.A. Heeren, F.A. Hamprecht:
    “Towards Digital Staining using Imaging Mass Spectrometry and Random Forests”
    Journal of Proteome Research, 8(7):3558-3567, 2009
    Abstract | BibTeX | PDF 

  • B. Menze, B. Kelm, N. Splitthoff, U. Köthe, F.A. Hamprecht:
    “On oblique random forests”
    in: Mach. Learning and Knowledge Discovery in Databases, Springer LNCS 6912, pp. 453-469, 2011.
    Abstract | PDF 

  • U. Köthe, F. Herrmannsdörfer, I. Kats, F.A. Hamprecht:
    “SimpleSTORM: a fast, self-calibrating reconstruction algorithm for localization microscopy”
    Histochemistry and Cell Biology, 141(6):613–627, 2014.
    Abstract | PDF 

  • H. Meine, U. Köthe, P. Stelldinger:
    “A Topological Sampling Theorem for Robust Boundary Reconstruction and Image Segmentation”
    Discrete Applied Mathematics (DGCI Special Issue), 157(3):524-541, 2009.
    Abstract | PDF 

  • U. Köthe: “What Can We Learn from Discrete Images about the Continuous World?”
    in: D. Coeurjolly, I. Sivignon, L. Tougne, F. Dupont (Eds.): Discrete Geometry for Computer Imagery, Proc. DGCI 2008, Springer LNCS 4992, pp. 4-19, 2008.
    Abstract | PDF 

  • P. Stelldinger, U. Köthe:
    “Towards a general sampling theory for shape preservation”
    Image and Vision Computing, Special Issue Discrete Geometry for Computer Vision, 23(2): 237-248, 2005.
    Abstract | PDF 

Curriculum Vitae

since 2018 Associate Professor and group leader in the Visual Learning Lab Heidelberg
26. Nov. 2008 Habilitation for a thesis entitled “Reliable Low-Level Image Analysis”, Department of Informatics, University of Hamburg
2008-2012 Senior scientist at Heidelberg Collaboratory for Image Processing (HCI)
2007-2017 Vice Group Leader of the Image Analysis ans Learning Group (formerly: Multidimensional Image Processing), University of Heidelberg
Spring semester 2004 Guest researcher at Computer Vision Laboratory, Linköping University, Sweden
1999-2007 Assistant professor (officially: “Hochschulassistent”) in the Cognitive Systems Group, University of Hamburg
29. Feb. 2000 Dr. rer. nat. (PhD) for a thesis entitled “Generische Programmierung für die Bildverarbeitung”, Department of Informatics, University of Hamburg
Spring semester 1993 Guest researcher at Sarnoff Corporation, Princeton, USA
1992-1999 Research assistant at the Fraunhofer Institute for Computer Graphics, Rostock
1986-1991 Study of physics at University of Rostock, Diploma thesis on “Mikroskopische Herleitung einer Ratengleichung am Beispiel der Nukleonen-Deuteron-Reaktion”