Multimedia Information Processing Group

Lars Schmarje

Hermann-Rodewald-Str. 3, R. 305 (HRS 3)
Phone: +49 431 880-1418
Telefax: +49 431 880-4845
las@informatik.uni-kiel.de

Short CV

  • since 2019
 PhD student, Kiel University, Germany
  • 2016 - 2018    
 MSc Computer Science, Kiel University, Germany
  Thesis: "Klassifizierung von Osteogenesis Imperfecta Strukturen in SHG-Scans, Methodenvergleich in 2D und 3D"
  • 2013 - 2016
 BSc Mathematics & Computer Science (Educational), Kiel University, Germany
  Thesis: "Entwicklung standardkonformer IT-Sicherheitskozepte"

Research topics

I am working on semi-supervised learning for image classification with deep neural networks. I'm investigating how we can decrease the needed supervision during training. Especially I'm interested in the usage of these methods in uncertain classification problems to improve the consistency between different persons and over time. For example this topic is highly valuable for medical image analysis when the used footage could be interpreted in numerous ways.

Publications

 

Lars Schmarje, Claudius Zelenka, Ulf Geisen, Claus-C. Glüer, Reinhard Koch
2D and 3D Segmentation of Uncertain Local Collagen Fiber Orientations in SHG Microscopy
41th German Conference on Pattern Recognition (DAGM GCPR 2019)
green open access   The final publication is available at Springer via https://doi.org/10.1007/978-3-030-33676-9_26 Preprint Dataset https://doi.org/10.5281/zenodo.3355937 Repository