Past and Running Thesis and Projects


  1. Exact Inference in Dense Graphical Models with Large Label Spaces
    Diplom Informatik, Stefan Haller, 10/2016
  2. Comparison of Learned Inference Approaches for Image Restoration
    Master Informatik, Jakob Kruse, 09/2016
  3. Improving 6D Pose Estimation from a Single RGB Frame by Hallucinating Depth
    Diplom Informatik, Eric Neumann, 02/2016
  4. Deep Convolutional Neural Network for Object Coordinate Regression
    Diplom Medieninformatik, Manuel Paternoster, 01/2016
  5. Semantic Segmentation of Topographic Plans
    Diplom Informatics, Daniel Schemala, 01/2016
  6. Casting Light on Object Pose Estimation via Object Coordinate Regression
    Master Informatik, Stephan Ihrke, 06/2015
  7. Enhanced depth estimation for “freehand stereo” using PatchMatch Stereo
    Master UASZ, Dominik Wetzel, 03/2016
  8. Discriminative Learning for Particle Filters in Micro-Biological Applications
    Master Informatik, Friedemann Pochert, 09/2015

Bachelor/Grosser Beleg

  1. Exact Inference in Graphical Models by Exploiting Partial Optimality and Shrinking Techniques
  2. Comparison of Different Decision Strategies for Conditional Random Fields
    Grosser Beleg, Aljoscha Fernández, 11/2015
  3. Statistical vs. Discriminative Learning of Convolutional Neural Networks for Semantic Labeling
    Grosser Beleg, Walter Forkel, Anatoly Zelenin, 10/2015
  4. Shadow Detection
    Bachelor, Steffen Temme, 03/2015
  5. Reduktion von Ausreißern bei Korrespondenzpunkten für die Stereorektifizierung
    Bachelor, Alexander Burkhardt, 10/2014

Lab Courses

  1. Comparison of Different CNN-Architectures
    Darja Arsentjeva, 09/2015
  2. Global Self Similarity
    Friedemann Pochert, 05/2015
  3. Super-Resolution with Regression Tree Fields
    Jakob Kruse, 05/2015
  4. Lighting Invariant Features for Decision Forest
    Xin Yang, 03/2015
  5. OpenGM Matching und Stereo
    Walter Forkel, 10/2014