Für eine korrekte Darstellung dieser Seite benötigen Sie einen XHTML-standardkonformen Browser, der die Darstellung von CSS-Dateien zulässt.


You are here

Visual Scene Analysis



In the witer semester 18/19 we will offer a new introductory course on deep learning methods, one of the hottest topics in image and data processing. The course will be 5 credits, counts towards the Computer Science Master for Visual Computing as well as the Mechatronics Master, and will be held in english. For more information about the course, we refer to http://www.vsa.informatik.uni-siegen.de/en/deep-learning

Additionally, we will offer an introductory course for numerical methods, i.e., learning about numerical algorithms that are crucial for the most fundamental tasks in data processing, such as solving linear equations, computing eigenvalues, or interpolating and integrating functions numerically. 

Open PhD/PostDoc position

We are always looking for highly motivated PhD candidates with an interest in numerical methods for Visual Computing and outstanding mathematical skills. If you are interested, please send an email to michael.moeller@uni-siegen.de, and include your CV as well as a letter describing your interest in the field. 


The visual scene analysis group conducts research in the field of mathematical image processing, computer vision, and machine learning with a special focus on the development of energy minimization methods for improving and analysing digital images. In this context we are interested in convex relaxations, functional lifting and multiscale methods as well as the development of efficient convex and nonconvex optimization algorithm for the solution of the corresponding energy minimization problems. Please visit our publications page to find out more about our recent research. 


Bachelor or Master Thesis

Feel free to contact me at michael.moeller@uni-siegen.de if you are interested in writing your Bachelor or Master thesis in the field of image processing or computer vision! We have several exciting topics! Please find a list of examples below. I am happy to explain these (sometimes technically sounding) topics in more detail. 

For instance

  • For scientific machine learning projects that contribute to the application of autonomous driving, we can offer master thesis (possibly even subsequent Ph.D. opportunities) with adptiv (https://www.aptiv.com/) in Wuppertal. Please contact me for further details. 
  • In deep learning, e.g. familiarize yourself with a deep learning framework such as http://pytorch.org/, and improve certain image reconstruction tasks, e.g. working on decompressing images with an architecture motivated by variational decompression methods. For the latter a particularly interesting interdisciplinary project could be a collaboration with the group of Prof. Weinberg (Festkörpermechanik) on the automatic analysis of open-pore polyurethane foam. A German description can be found here
  • Tackling image or video reconstruction tasks with deep learning, with the goal to train "image improvement networks" that are (partially) oblivious to the type of distortion that ought to be removed. (In collaboration with the computer graphics group).  
  • In 3D reconstruction, e.g. familiarize yourself with the relation of images and the camera pose to real world coordinates and try to reimplement an energy minimization approach to surface reconstruction similar to https://www.youtube.com/watch?v=TGg-ujjSsOM (but not in real-time). Extensions of such techniques are able to map large scale geometric features, see e.g. https://www.youtube.com/watch?v=GnuQzP3gty4
  • In optimization, e.g.
    • on joint optimization methods for biconvex or nonconvex problems such as blind deblurring or super resolution. Some suprising results show that there is a gap between good results and faithful optimization (http://www.cvg.unibe.ch/dperrone/tvdb/), which could be investigated in more detail!
    • on optimizing deeply nested functions (as frequently arising in deep learning applications) with proximal splitting methods.
    • on efficient ways to optimize a certain class of constrained convex functions whose proximal operator does not have a closed form.
    • on optimizing nonconvex functions by representing the underlying problem in a higher dimensional space, also known as lifting. See https://arxiv.org/pdf/1512.01383.pdf for an example of a recent work on this topic. I can give clear instructions of how a project on a related topic can look like. 
  • In combining optimization and deep learning methods, e.g. by studying and extending our recent work on using denoising networks as proximal operators, see https://arxiv.org/abs/1704.03488.
  • On multiscale methods for inverse problems, extending the current theory on nonlinear spectral decompositions (see, e.g. https://arxiv.org/pdf/1510.01077.pdf). Such a project could be application driven (https://arxiv.org/pdf/1703.08001.pdf), or based on some mathematical analysis of generalized eigenfunctions, see https://arxiv.org/pdf/1601.02912v1.pdf

If you have ideas for your own research project, let me know!