University Freiburg

In recent years, Prof. Brox's Department of Pattern Recognition and Image Processing has very successfully contributed to the establishment of deep learning in the field of computer vision and has published numerous internationally renowned papers, for example on unsupervised learning of invariant features, learning of 3D object representations, use of synthetic training data, semantic segmentation, as well as the establishment of depth, stereo and motion estimation as learning problems.

The working group is an international leader in learning methods for 3D perception. The current research aims especially at the prediction of probability distributions and sequential learning.

With this expertise, the University of Freiburg will in particular contribute new solutions to challenges in the areas of robustness in Open World, continuous learning, partially supervised and unsupervised learning and active learning.