In this theme we look at optimization and learning in directed and undirected graphical models. undirected graphical models have factors that depend on many variables. This includes higher-order models as well as models with a large, even continuous, label space. Optimizing in such models is often NP-hard. Furthermore, it is often difficult to hand-craft the functional relationship between variables, hence it is necessary to learn them. The goal in this theme is to analyse the trade-offs between models, optimization and learning with the ultimate goal of achieving practically relevant algorithms which are efficient and accurate. A shortlist of research topics we are excited about: a) optimization in undirected graphical models with higher-order factors, continuous label space, and models of very large size; b) combining generative and discriminative models; c) probabilistic learning and inference in undirected graphical models; d) combining deep directed models with undirected graphical models.
- Diverse Solutions to Graphical Models
- A link to the Open GM 2 project where we have been involved in
Recent (selected) articles:
- J. Kappes, B. Andres, F. Hamprecht, C. Schnoerr, S. Nowozin, D. Batra, S. Kim, B. Kausler, J. Lellmann, N. Komodakis, and C. Rother, A Comparative Study of Modern Inference Techniques for Discrete Energy Minimization Problems, CVPR 2013
- J. Jancsary, S. Nowozin, T. Sharp, and C. Rother, Regression Tree Fields – An Efficient, Non-parametric Approach to Image Labeling Problems, CVPR 2012.
- V. Lempitsky, A. Blake, and C. Rother, Branch-and-Mincut: Global Optimization for Image Segmentation with High-Level Priors, in Journal of Mathematical Imaging and Vision (JMIV), 2012
- P. Kohli and C. Rother, Higher-Order Models in Computer Vision, in Image Processing and Analysis with Graphs, CRC Press, 2012
- V. Kolmogorov and C. Rother, Minimizing non-submodular functions with graph cuts – a review, PAMI,vol. 29, no. 7, 2007.