Home | deutsch  | Legals | Data Protection | Sitemap | KIT

Contact Information
Karlsruhe Institute of Technology
Vision and Fusion Laboratory (IES)

Prof. Dr.- Ing. Jürgen Beyerer
c/o Technologiefabrik
Haid-und-Neu-Str. 7
76131 Karlsruhe

Tel:  +49 721 - 608 45910
Fax: +49 721 - 608 45926

Welcome to Vision and Fusion Laboratory (IES)

 

Prof. Dr.-Ing. J. Beyerer

 

Latest

Klausur Pattern Recognition

The next written exam (Klausur) in Pattern Recognition (Mustererkennung) will take place on 5th August 2019.

  • Beginning of online registration period: 15th May 2019
  • End of online registration period: 19th July 2019
  • Beginning of signing out: 15th May 2019
  • End of signing out: 26th July 2019

Klausur Automated Visual Inspection and Image Processing

Concerning the lecture Automated Visual Inspection and Image Processing until 30th September 2019 it will still be possible to do an oral exam.
Mrs. Gross can inform you about possible exam days.
From WS 19/20 onwards there will be a written exam (Klausur) in ASB.
The first “Klausur” will take place around February 2020. The exact date will be announced on time.

Lectures offered at the Chair can be found here.

Further information on Students' Theses can be found here.

 
 
Maschinelles Lernen: Low-Dimensional Embeddings and Topology
Typ:

Hiwi-Stelle

Betreuer:

Tim Zander

Status:

laufende Arbeit

Low-dimensional embeddings, also called manifold learning, are a central dimension reduction technique in data analysis. It is typically used for visual data inspection as well as a preprocessing step in a data pipeline.

The source (Rieck 2017)[Ch. 7] evaluates a range of classical embedding algorithms with typical embedding quality measures. In this project, the recent UMAP (McInnes and Healy 2018) and other older algorithms, which our group is currently implementing, shall be added to the analysis. Distances between Persistence Diagrams (PD) from the original and the embedded data shall be investigated as a new quality measure for low-dimensional embeddings. PDs are able to capture topological information, like circles and spheres, in the learned manifold. Such a scale-free local-to-global structure puts PDs into an interesting position within the other quality measures.

Possible Python APIs to the algorithms (many are actually written in C++) can be found here:

Applicants should not be afraid of a mathematical approach to programming. The project is supervised by Tim Zander (KIT, ) with cooperation from Arkadi Schelling (Uni Bremen, ).

McInnes, Leland, and John Healy. 2018. “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction.” ArXiv E-Prints, February. http://arxiv.org/abs/1802.03426.

Rieck, Bastian Alexander. 2017. “Persistent Homology in Multivariate Data Visualization.” PhD thesis, University of Heidelberg.