This is an old revision of the document!
Research Interests
Variational Methods, Convex Optimization, Operator Splitting Algorithms, GPU Programming
Brief Bio
Thomas Möllenhoff received bis Bachelor's degree (2011) and his Master's degree (2014, with high distinction) in Computer Science from the Technical University of Munich (Germany). In 2013 he studied one semester abroad at the Technical University of Denmark, Copenhagen. Since February 2014 he is a PhD Student in the Computer Vision Group at the Technical University of Munich, Germany, headed by Professor Daniel Cremers.
Publications
List of publications.
Export as PDF, XML, TEX or BIB
Journal Articles
2022
[] A Cutting-Plane Method for Sublabel-Accurate Relaxation of Problems with Product Label Spaces , In International Journal of Computer Vision (IJCV), 2022. ([code])
[] Lifting the Convex Conjugate in Lagrangian Relaxations: A Tractable Approach for Continuous Markov Random Fields , In SIAM J. Imaging Sci., volume 15, 2022.
2015
[] The Primal-Dual Hybrid Gradient Method for Semiconvex Splittings , In SIAM Journal on Imaging Sciences, volume 8, 2015.
Conference and Workshop Papers
2024
[] Variational Learning is Effective for Large Deep Networks , In International Conference on Machine Learning (ICML), 2024. ([code][blog][tutorial])
Spotlight
2021
[] Sublabel-Accurate Multilabeling Meets Product Label Spaces , In DAGM German Conference on Pattern Recognition (GCPR), 2021. ([presentation] [code])
Oral Presentation
2020
[] Optimization of Graph Total Variation via Active-Set-based Combinatorial Reconditioning , In International Conference on Artificial Intelligence and Statistics (AISTATS), 2020. ([code])
2019
[] Informative GANs via Structured Regularization of Optimal Transport , In NeurIPS Workshop on Optimal Transport and Machine Learning, 2019.
[] Controlling Neural Networks via Energy Dissipation , In International Conference on Computer Vision (ICCV), 2019.
[] Flat Metric Minimization with Applications in Generative Modeling , In International Conference on Machine Learning (ICML), 2019. (arXiv:1905.04730, code, talk)
Full Oral Presentation [] Lifting Vectorial Variational Problems: A Natural Formulation based on Geometric Measure Theory and Discrete Exterior Calculus , In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. (arXiv:1905.00851, talk)
Oral Presentation
2018
[] Proximal Backpropagation , In International Conference on Learning Representations (ICLR), 2018. (arXiv:1706.04638, code)
[] Combinatorial Preconditioners for Proximal Algorithms on Graphs , In International Conference on Artificial Intelligence and Statistics (AISTATS), 2018.
[] Fight ill-posedness with ill-posedness: Single-shot variational depth super-resolution from shading , In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2018. ([supp] [poster] [slides] [code] [cvf] [video])
Spotlight Presentation
2017
[] Sublabel-Accurate Discretization of Nonconvex Free-Discontinuity Problems , In International Conference on Computer Vision (ICCV), 2017. ([supp])
2016
[] Sublabel-Accurate Convex Relaxation of Vectorial Multilabel Energies , In European Conference on Computer Vision (ECCV), 2016. ([supp] [code])
[] Sublabel-Accurate Relaxation of Nonconvex Energies , In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. ([supp] [code])
Oral Presentation, Received the Best Paper Honorable Mention Award at CVPR 2016
2015
[] Low Rank Priors for Color Image Regularization , In Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR), 2015.
2013
[] Efficient Convex Optimization for Minimal Partition Problems with Volume Constraints , In Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR), 2013.