Prof. Dr. Manuel Günther

Portrait of Prof. Manuel Günther

 

 

Prof. Dr. Manuel Günther

Artificial Intelligence and Machine Learning Group
Department of Informatics
University of Zurich
Andreasstr. 15 / Office AND 2.54
8050 Zürich
Switzerland

Phone: +41 44 635 71 40

E-Mail: guenther@ifi.uzh.ch

(not so short) Biography

Prof. Dr. Manuel Günther started his position as Assistant Professor for Artificial Intelligence and Machine Learning at the Department of Informatics of the University of Zurich in July 2020.

His academic career went through several different locations. He received his diploma in Computer Science with a major subject of machine learning from the Technical University of Ilmenau, Germany, in 2004. His doctoral thesis was written in the Institut für Neuroinformatik at the Ruhr University of Bochum, Germany, between 2004 and 2011 about statistical extensions of Gabor graph based face detection, recognition and classification techniques (PDF, 12115 KB). Finally, he received his doctoral degree (Dr.-Ing.) from the Technical University of Ilmenau in 2012.

Between 2012 and 2015, Prof. Günther was a postdoctoral researcher in the Biometrics Group at the Idiap Research Institute in Martigny, Switzerland. Since then, he is actively participating in the implementation of the open source signal processing and machine learning library Bob. Particularly, he was the leading developer of the Biometric Recognition packages, a library to run biometric recognition experiments, which he presented as a hands-on tutorial at the International Joint Conference on Biometrics (IJCB), 2017, recordings of which are on YouTube.

From 2015 to 2018, Prof. Günther was working as a Research Associate at the Vision and Security Technology Lab at the University of Colorado Colorado Springs, Colorado, USA. There, he developed algorithms for the alignment-free classification of facial attributes from single images. His research also included the classification of samples under the presence of unknown classes. Particularly, he developed algorithms for the open-set identification of human faces, which he applied in two Unconstrained Face Detection and Open Set Recognition Challenges, which he was leading himself.

After a short industry excursion at the trinamiX GmbH in Ludwigshafen, Germany, Prof. Günther accepted a call as Assistant Professor at the Department of Informatics. There, he is continuing his research on the classification  of the unknown and on automatic face recognition. His research interests also include deep learning in general and the phenomenon of adversarial samples in particular. Furthermore, he promotes the idea of Reproducible Research, which allows researchers to start their work at the state of the art.

Publications

2020

  • Akshay Raj Dhamija, Manuel Günther, Jonathon Ventura and Terrance E. Boult. The Overlooked Elephant of Object Detection: Open Set. Winter Conference on Applications in Computer Vision (WACV), 2020. pdf; source code
  • Manuel Günther, Akshay Raj Dhamija and Terrance E. Boult. Watchlist Adaptation: Protecting the Innocent. International Conference of the Biometrics Special Interest Group (BIOSIG), 2020. pdf

2019

  • Terrance E. Boult, Steve Cruz, Akshay Raj Dhamija, Manuel Günther, James Henrydoss and Walter J. Scheirer. Learning and the Unknown: Surveying Steps toward Open World Recognition.  AAAI Conference on Artificial Intelligence, 2019. pdf
  • Andras Rozsa, Manuel Günther, Ethan M. Rudd and Terrance E. Boult. Facial Attributes: Accuracy and Adversarial Robustness. Pattern Recognition Letters, 2019. arXiv ID 1801.02480
  • Akshay Raj Dhamija, Manuel Günther and Terrance E. Boult. Improving Deep Network Robustness to Unknown Inputs with Objectosphere. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019. pdf

2018

  • Akshay Raj Dhamija, Manuel Günther and Terrance E. Boult. Reducing Network Agnostophobia. Advances in Neural Information Processing Systems (NIPS), 2018. pdf; source code
  • Chunchun Li, Manuel Günther and Terrance E. Boult. ECLIPSE: Ensembles of Centroids Leveraging Iteratively Processed Spatial Eclipse Clustering. Winter Conference on Applications in Computer Vision (WACV), 2018.
  • Andras Rozsa, Manuel Günther and Terrance E. Boult. Towards Robust Deep Neural Networks With BANG. Winter Conference on Applications in Computer Vision (WACV), 2018. arXiv ID 1612.00138
  • Michael Bihn, Manuel Günther, Daniel Lemmond and Terrance E. Boult. Evaluating a Convolutional Neural Network on ShortWave Infra-Red Images.Winter Conference on Applications in Computer Vision (WACV) Cross-Domain Face Recognition Workshop, 2018.
  • Ethan M. Rudd, Manuel Günther, Akshay Raj Dhamija, Faris A. Kateb and Terrance E. Boult. What's Hiding in My Deep Features? Deep Learning in Biometrics, CRC Press, 2018. pdf

2017

  • Manuel Günther, Andras Rozsa and Terrance E. Boult. AFFACT: Alignment-Free Facial Attribute Classification Technique. International Joint Conference on Biometrics (IJCB), 2017. arXiv ID 1611.06158Trained Caffe networks
  • Manuel Günther and others. Unconstrained Face Detection and Open-Set Face Recognition Challenge. International Joint Conference on Biometrics (IJCB), 2017. arXiv ID 1708.02337competition website
  • Andras Rozsa, Manuel Günther and Terrance E. Boult. LOTS about Attacking Deep Features. International Joint Conference on Biometrics (IJCB), 2017. arXiv ID 1611.06179
  • André Anjos, Manuel Günther, Tiago de Freitas Pereira, Pavel Korshunov, Amir Mohammadi and Sébastien Marcel. Continuously Reproducing Toolchains in Pattern Recognition and Machine Learning Experiments. International Conference on Machine Learning (ICML) Workshop on Reproducibility in Machine Learning, 2017. pdfsource code
  • Andras Rozsa, Manuel Günther and Terrance E. Boult. Adversarial Robustness: Softmax versus Openmax. British Machine Vision Conference (BMVC), 2017. arXiv ID 1708.01697
  • Manuel Günther, Steve Cruz, Ethan M. Rudd and Terrance E. Boult. Toward Open-Set Face Recognition. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2017. arXiv ID 1705.01567
  • James Henrydoss, Steve Cruz, Ethan M. Rudd, Manuel Günther and Terrance E. Boult. Incremental Open Set Intrusion Recognition Using Extreme Value Machine. IEEE International Conference on Machine Learning and Applications (ICMLA), 2017. Won Best Poster award.
  • Ethan M. Rudd, Andras Rozsa, Manuel Günther and Terrance E. Boult. A Survey of Stealth Malware: Attacks, Mitigation Measures, and Steps Toward Autonomous Open World Solutions. IEEE Communications Surveys & Tutorials, 2017. arXiv ID 1603.06028

2016

  • Andras Rozsa, Manuel Günther and Terrance E. Boult. Are Accuracy and Robustness Correlated? IEEE International Conference on Machine Learning and Applications (ICMLA), 2016. arXiv ID 1610.04563
  • Andras Rozsa, Manuel Günther, Ethan M. Rudd and Terrance E. Boult. Are Facial Attributes Adversarially Robust? International Conference on Pattern Recognition (ICPR), 2016. arXiv ID 1605.05411 Won Best Student Paper award.
  • Ethan M. Rudd, Manuel Günther and Terrance E. Boult. MOON: A Mixed Objective Optimization Network for the Recognition of Facial Attributes. European Conference on Computer Vision (ECCV), 2016. arXiv ID 1603.07027Trained Caffe networks
  • Ethan M. Rudd, Manuel Günther and Terrance E. Boult. PARAPH: Presentation Attack Rejection by Analyzing Polarization Hypotheses. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2016. arXiv ID 1605.03124
  • Manuel Günther, Laurent El Shafey and Sébastien Marcel. Face Recognition in Challenging Environments: An Experimental and Reproducible Research Survey. Face Recognition Across the Imaging Spectrum, Springer, 2016. pdfsource code

Earlier

  • Abhishek Dutta, Manuel Günther, Laurent El Shafey, Sébastien Marcel, Raymond Veldhuis and Luuk Spreeuwers. Impact of Eye Detection Error on Face Recognition Performance. IET Biometrics, 2015. pdfsource code
  • Manuel Günther, Stefan Böhringer, Dagmar Wieczorek and Rolf P. Würtz. Reconstruction of Images from Gabor Graphs with Applications in Facial Image Processing. Journal of Wavelets, Multiresolution and Information Processing, 2015. pdf
  • Rakesh Metha, Manuel Günther and Sübastien Marcel. Gender Classification by LUT based Boosting of Overlapping Block Patterns. Scandinavian Conference on Image Analysis (SCIA), 2015. pdfsource code
  • Miranti I. Mandasari, Manuel Günther, Roy Wallace, Rahim Saedi, Sébastien Marcel and David Van Leeuwen. Score Calibration in Face Recognition. IET Biometrics, 2014. pdfsource code
  • Elie Khoury, Laurent El Shafey, Chris McCool, Manuel Günther and Sébastien Marcel. Bi-modal Biometric Authentication on Mobile Phones in Challenging Conditions. Image and Vision Computing (IVC), 2014. pdf
  • Elie Khoury, Manuel Günther, Laurent El Shafey and Sébastien Marcel. On the Improvements of Uni-modal and Bi-modal Fusions of Speaker and Face Recognition for Mobile Biometrics. Biometric Technologies in Forensic Science (BTFS), 2013. pdfsource code
  • Manuel Günther and others. The 2013 Face Recognition Evaluation in Mobile Environment. International Conference on Biometrics (ICB), 2013. pdf
  • Elie Khoury and others. The 2013 Speaker Recognition Evaluation in Mobile Environment. International Conference on Biometrics (ICB), 2013. pdf
  • Manuel Günther, Roy Wallace and Sébastien Marcel. An Open Source Framework for Standardized Comparisons of Face Recognition Algorithms. European Conference on Computer Vision (ECCV), Workshops and Demonstrations: 547-556, 2012. pdfsource code
  • André Anjos, Laurent El Shafey, Roy Wallace, Manuel Günther, Chris McCool and Sébastien Marcel. Bob: a Free Signal Processing and Machine Learning Toolbox for Researchers. ACM Multimedia Conference, 2012. pdfhome page
  • Manuel Günther, Dennis Haufe and Rolf P. Würtz. Face Recognition with Disparity Corrected Gabor Phase Differences. Artificial Neural Networks and Machine Learning (ICANN), 2012. pdf
  • Manuel Günther. Statistical Gabor Graph Based Techniques for the Detection, Recognition, Classification, and Visualization of Human Faces. PhD thesis, Institut für Neuroinformatik, Technische Universität Ilmenau, 2011. pdf (PDF, 12115 KB)
  • Harald J. Schneider, Robert P. Kosilek, Manuel Günther, J. Römmler., G.K. Stalla, C. Sievers, M. Reincke, Jochen Schopohl and Rolf P. Würtz. A novel approach to the detection of acromegaly: accuracy of diagnosis by automatic face classification. Journal of Clinical Endocrinology and Metabolism, 2011.
  • Stefan Böhringer and others. Genetic determination of human facial morphology: links between cleft-lips and normal variation. European Journal of Human Genetics, 2011. pdf
  • Stefan Böhringer, Manuel Günther, Stella Sinigerova, Rolf P. Würtz, Bernard Horsthemke and Dagmar Wieczorek. Automated Syndrome Detection in a set of Clinical Facial Photographs. American Journal of Medical Genetics, 2011.
  • Manuel Günther, Marco K. Müller and Rolf P. Würtz. Two kinds of Statistics for Better Face Recognition. International Conference on Numerical Analysis and Applied Mathematics (ICNAAM), 2010. pdf
  • Manuel Günther and Rolf P. Würtz. Face detection and recognition using maximum likelihood classifiers on Gabor graphs. International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI), 2009. pdf
  • Manuel Günther. Klassifikation von Gesichtern mit optimierten lokalen Graphen auf 2D und 3D Bilddaten. Diploma thesis, Technische Universität Ilmenau, 2005. pdf (German only) (PDF, 11754 KB)