Visual Inference
The Visual Inference Lab at TU Darmstadt, led by Prof. Stefan Roth, conducts research in several areas of computer vision with an emphasis on statistical methods and machine learning. We develop mathematical models and algorithms for analyzing and processing digital images with the computer. For example, we work on semantic scene understanding, image motion estimation, deep learning, probabilistic methods, image restoration, and object tracking.
Moreover, we regularly offer courses, seminars and labs in computer science, particularly in computer vision and machine learning.
Recent Highlight
News & Events
-
Limited Reachability
Due to the current situation caused by COVID-19, the VISINF team is working from home. We are best reached by email. Stay healthy!
-
Available Bachelor and Master Thesis Topics in Visual Inference Lab
A detailed list of all available master and bachelor thesis topics are accessible here. Please note that you need to login using your TU-ID and password to be able to view this page.
-
PhD Defense
January 13, 2020
Stephan Richter has successfully defended his PhD thesis “Visual Perception with Synthetic Data”. Congratulations!
-
ERC Consolidator Grant
December 10, 2019
Stefan Roth has been awarded a Consolidator Grant from the European Research Council (ERC) for his project “RED – Robust, Explainable Deep Networks in Computer Vision.” [press release]
-
CVPR 2019 Outstanding Reviewers
June 9, 2019
Jochen Gast, Junhwa Hur, and Anne Wannenwetsch were chosen as outstanding reviewers for CVPR 2019. Congratulations!
Recent Publications
-
CVPR 2020
A. S. Wannenwetsch and S. Roth, “Probabilistic pixel-adaptive refinement networks.” [preprint], [code]
-
CVPR 2020
J. Hur and S. Roth, “Self-supervised monocular scene flow estimation,” oral presentation. [preprint], [code]
-
CVPR 2020
N. Araslanov and S. Roth, “Single-stage semantic segmentation from image labels.” [coming soon]
-
ICLR 2020
S. Mahajan, I. Gurevych, and S. Roth, “Latent normalizing flows for many-to-many cross domain mappings.” [paper], [code]