Stochastic normalizing flows for inverse problems: a Markov Chains viewpoint
To overcome topological constraints and improve the expressiveness of normalizing flow
architectures, Wu, Köhler, and Noé introduced stochastic normalizing flows which combine …
architectures, Wu, Köhler, and Noé introduced stochastic normalizing flows which combine …
Geometric numerical integration of the assignment flow
The assignment flow is a smooth dynamical system that evolves on an elementary statistical
manifold and performs contextual data labeling on a graph. We derive and introduce the …
manifold and performs contextual data labeling on a graph. We derive and introduce the …
Assignment flows for data labeling on graphs: convergence and stability
A Zern, A Zeilmann, C Schnörr - Information Geometry, 2022 - Springer
The assignment flow recently introduced in the J. Math. Imaging and Vision 58/2 (2017)
constitutes a high-dimensional dynamical system that evolves on a statistical product …
constitutes a high-dimensional dynamical system that evolves on a statistical product …
Recent advances in denoising of manifold-valued images
R Bergmann, F Laus, J Persch, G Steidl - Handbook of Numerical Analysis, 2019 - Elsevier
Modern signal and image acquisition systems are able to capture data that are no longer
real-valued but may take values on a manifold. However, whenever measurements are …
real-valued but may take values on a manifold. However, whenever measurements are …
A convergent iterative support shrinking algorithm for non-lipschitz multi-phase image labeling model
Y Yang, Y Li, C Wu, Y Duan - Journal of Scientific Computing, 2023 - Springer
The non-Lipschitz piecewise constant Mumford–Shah model has been shown effective for
image labeling and segmentation problems, where the non-Lipschitz isotropic ℓ p (0< p< 1) …
image labeling and segmentation problems, where the non-Lipschitz isotropic ℓ p (0< p< 1) …
A nonlocal denoising algorithm for manifold-valued images using second order statistics
F Laus, M Nikolova, J Persch, G Steidl - SIAM Journal on Imaging Sciences, 2017 - SIAM
Nonlocal patch-based methods, in particular the Bayesian approach of Lebrun, Buades, and
Morel [SIAM J. Imaging Sci., 6 (2013), pp. 1665--1688], are considered to be state-of-the-art …
Morel [SIAM J. Imaging Sci., 6 (2013), pp. 1665--1688], are considered to be state-of-the-art …
A graph framework for manifold-valued data
R Bergmann, D Tenbrinck - SIAM Journal on Imaging Sciences, 2018 - SIAM
Graph-based methods have been proposed as a unified framework for discrete calculus of
local and nonlocal image processing methods in recent years. In order to translate …
local and nonlocal image processing methods in recent years. In order to translate …
Image labeling based on graphical models using wasserstein messages and geometric assignment
R Hühnerbein, F Savarino, F Åström… - SIAM Journal on Imaging …, 2018 - SIAM
We introduce a novel approach to Maximum A Posteriori (MAP) inference based on discrete
graphical models. By utilizing local Wasserstein distances for coupling assignment …
graphical models. By utilizing local Wasserstein distances for coupling assignment …
Unsupervised assignment flow: Label learning on feature manifolds by spatially regularized geometric assignment
This paper introduces the unsupervised assignment flow that couples the assignment flow
for supervised image labeling (Åström et al. in J Math Imaging Vis 58 (2): 211–238, 2017) …
for supervised image labeling (Åström et al. in J Math Imaging Vis 58 (2): 211–238, 2017) …
An image registration model in electron backscatter diffraction
Recently, variational methods were successfully applied for computing the optical flow in
gray and RGB-valued image sequences. A crucial assumption in these models is that pixel …
gray and RGB-valued image sequences. A crucial assumption in these models is that pixel …