Jump label

Service navigation

Main navigation

You are here:

Main content

Theses

The following list should give an idea of possible topics and current fields of research at the Pattern Recognition group. Generally, we will very rarely provide readily specified topics, but will try to define them in close collaboration with our students. For a list of completed and ongoing theses, please refer to our publications page.

Most topics are suitable for both Bachelor and Master/Diploma theses, but requirements and complexity will be adjusted accordingly. Despite the fact, that all of the following topic descriptions are in English, candidates are free to choose between German and English for writing their actual thesis.

Theses Topics

Document Analysis/Handwriting Recognition

Human Activity Recognition and Face Recognition

    • Attribute CNNs for Human Activity Recognition

    • Semi-Automated Annotation of Motion Capturing Data

    • Face Recognition and Identification

    • Synthetic Data for Face Recognition and Generation

    • Self-Supervised Learning for Activity Recognition ([1], [2])
    • Human Identification using Multi-channel Time-Series

Explainable Artificial Intelligence for HAR

... and more, contact Nilah Ravi Nair

Semantic Segmentation

  • Segmentation in Remote Sensing

  • Deep Semantic Segmentation

  • Unsupervised Segmentation Models

  • Object Detection and Recognition

  • ... and more, contact Dominik Koßmann

Deep Learning and Neural Networks

Many other interesting topics in the fields of human-computer-interaction and pattern recognition are available. Please contact the staff for further information.

Requirements

If you are interested in working with us, you should fulfill several of the following requirements:

  • Own initiative, creativity, ability to work in a team: You will get involved in interesting (and challenging) research projects.
  • Good programming abilities (Python, C/C++, Java, Shell programming, ...)
  • Experience in the usage of and development under UNIX (Linux).
  • Previous knowledge in Pattern Recognition, Computer Vision etc., e.g. by attending the appropriate lectures.