Institute of Computer Graphics and Vision

Team Pock

Vision, Learning and Optimization Group

Within the fruitful research environment of the Institute for Computer Graphics and Vision, the focus of our research is the development of mathematical models for computer vision and image processing in mobile scenarios as well as the development of efficient algorithms to compute these models. Our group consists currently of 10 PhD students and 2 PostDocs. Find out more about our research, our team and published software!

"Nothing is more practical than a good theory."
Kurt Lewin

News

October 2021. New publication by Marc Masana "On the importance of cross-task features for class-incremental learning" [paper] [code] Accepted at the Theory and Foundation of Continual Learning Workshop, International Conference on Machine Learning (ICML) 2021.
September 2020. Thomas Pock is giving a talk on "Variation modeling meets learning" in the Imaging & Inverse Problems (IMAGINE) One World seminars [website]
July 2020. New paper by Thomas Grandits et al. on "An inverse Eikonal method for identifying ventricular activation sequences from epicardial activation maps" [paper]
July 2020. Thomas Pock is giving an invited plenary talk on Variational Networks at the (virtual) SIAM Conference on Imaging Science.
June 2020. New preprint by Erich Kobler et al. on "Total Deep Variation: A Stable Regularizer for Inverse Problems" [arXiv]
June 2020. Check out our CVPR 2020 teaser song on Total Deep Variation on our YouTube channel
March 2020. New preprint by Patrick Knöbelreiter et al. on "Belief Propagation Reloaded: Learning BP-Layers for Labeling Problems" [arXiv]
March 2020. We joined Twitter! Follow our account @VLOgroupGraz for updates on our research
February 2020. Thomas Pock is visiting École Polytechnique to give a lecture series on "Vision, Optimization and Learning" [course material]
January 2020. New preprint by Erich Kobler et al. on "Total Deep Variation for Linear Inverse Problems" [arXiv]
December 2019. New preprint by Markus Hofinger et al. on "The Five Elements of Flow" [arXiv]
September 2019. Patrick Knöbelreiter and Thomas Pock received the GCPR Best Paper Award for "Learned Collaborative Stereo Refinement" [arXiv]

To top

Team Menu

	Institute for Computer Graphics and Vision
	Graz University of Technology Inffeldgasse 16/II 8010 Graz, Austria
	mail: pocknoSpam@icg.tugraz.at phone: +43 316 873-5056

By loading the Tweets your data will be transmitted to Twitter. The privacy policy of Twitter shall apply.

Vision, Learning and Optimization Group

Tweets by ‎@VLOgroupGraz