An Efficient Fusion Move Algorithm for the Minimum Cost Lifted Multicut Problem

  title={An Efficient Fusion Move Algorithm for the Minimum Cost Lifted Multicut Problem},
  author={Thorsten Beier and Bj{\"o}rn Andres and U. K{\"o}the and Fred A. Hamprecht},
Many computer vision problems can be cast as an optimization problem whose feasible solutions are decompositions of a graph. [] Key Method We propose a fusion move algorithm for computing feasible solutions, better and more efficiently than existing algorithms. We demonstrate this and applications to image segmentation, obtaining a new state of the art for a problem in biological image analysis.
Solving Minimum Cost Lifted Multicut Problems by Node Agglomeration
This work proposes two variants of a heuristic solver (primal feasible heuristic), which greedily generate solutions within a bounded amount of time for the minimum cost lifted multicut problem.
Higher-Order Multicuts for Geometric Model Fitting and Motion Segmentation.
This paper proposes a pseudo-boolean formulation for a multiple model fitting problem based on a formulation of any-order minimum cost lifted multicuts, which allows to partition an undirected graph with pairwise connectivity such as to minimize costs defined over any set of hyper-edges.
The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning
The Mutex Watershed is proposed, an efficient algorithm for graph partitioning that can accommodate not only attractive but also repulsive cues, allowing it to find a previously unspecified number of segments without the need for explicit seeds or a tunable threshold.
Higher-Order Minimum Cost Lifted Multicuts for Motion Segmentation
  • Margret Keuper
  • Computer Science
    2017 IEEE International Conference on Computer Vision (ICCV)
  • 2017
This paper introduces a generalization of the minimum cost lifted multicut problem to hypergraphs, and proposes a simple primal feasible heuristic that allows for a reasonably efficient inference in instances of higher-order liftedMulticut problem instances defined on point trajectory hyper graphs for motion segmentation.
A Benders Decomposition Approach to Correlation Clustering
  • Jovita Lukasik, Margret Keuper, M. Singh, Julian Yarkony
  • Computer Science
    2020 IEEE/ACM Workshop on Machine Learning in High Performance Computing Environments (MLHPC) and Workshop on Artificial Intelligence and Machine Learning for Scientific Applications (AI4S)
  • 2020
The Benders decomposition approach provides a promising new avenue to accelerate optimization for CC, and, in contrast to previous cutting plane approaches, theoretically allows for massive parallelization.
The Mutex Watershed: Efficient, Parameter-Free Image Partitioning
When presented with short-range attractive and long-range repulsive cues from a deep neural network, the Mutex Watershed gives results that currently define the state-of-the-art in the competitive ISBI 2012 EM segmentation benchmark.
Massively Parallel Benders Decomposition for Correlation Clustering
The Benders decomposition approach provides a promising new avenue to accelerate optimization for CC, and allows for massive parallelization.
RAMA: A Rapid Multicut Algorithm on GPU
This work proposes a highly parallel primal-dual algorithm for the multicut problem, a classical graph clustering problem widely used in machine learning and computer vision, and shows resulting one to two orders-of-magnitudes improvements in execution speed without sac-rificing solution quality compared to traditional sequential algorithms that run on CPUs.
Analysis and Optimization of Graph Decompositions by Lifted Multicuts
To find optimal decompositions defined by minimum cost lifted multicuts, this work establishes some properties of some facets of lifted multicut polytopes, define efficient separation procedures and apply these in a branch-and-cut algorithm.
Learning to solve Minimum Cost Multicuts efficiently using Edge-Weighted Graph Convolutional Neural Networks
This work employs a reformulation of the multicut ILP constraints to a polynomial program as loss function that allows to learn feasible multicut solutions in a scalable way and provides the first approach towards end-to-end trainable multicuts.


Efficient Decomposition of Image and Mesh Graphs by Lifted Multicuts
This work proposes a generalization of the MP with long-range terms (LMP) and designs and implements two efficient algorithms (primal feasible heuristics) for the MP and LMP which allow us to study instances of both problems w.r.t. the pixel grid graphs of the images in the BSDS-500 benchmark.
Higher-order segmentation via multicuts
Cut, Glue, & Cut: A Fast, Approximate Solver for Multicut Partitioning
The proposed algorithm finds segmentations that, as measured by a loss function, are as close to the ground-truth as the global optimum found by exact solvers, which is important for large-scale problems.
Fast Planar Correlation Clustering for Image Segmentation
This work describes a new optimization scheme for finding high-quality clusterings in planar graphs that uses weighted perfect matching as a subroutine and demonstrates this approach on the problem of image segmentation where this approach outperforms existing global optimization techniques in minimizing the objective.
Variable grouping for energy minimization
An energy-aware method for merging random variables to reduce the size of the energy to be minimized and a number of extremely efficient variable grouping strategies are proposed and evaluated.
Fusion Moves for Markov Random Field Optimization
This paper demonstrates one possible way of using graph cuts to combine pairs of suboptimal labelings or solutions, and proposes new optimization schemes for computer vision MRFs with applications to image restoration, stereo, and optical flow, among others.
A cutting plane algorithm for a clustering problem
This paper describes a cutting plane algorithm that is based on the simplex method and uses exact and heuristic separation routines for some of the classes of facets of the associated polytope.
Fusion moves for correlation clustering
This algorithm iteratively fuses the current and a proposed partitioning which monotonously improves the partitioning and maintains a valid partitioning at all times.
What energy functions can be minimized via graph cuts?
  • V. Kolmogorov, R. Zabih
  • Computer Science, Mathematics
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2004
This work gives a precise characterization of what energy functions can be minimized using graph cuts, among the energy functions that can be written as a sum of terms containing three or fewer binary variables.
Ensemble Segmentation Using Efficient Integer Linear Programming
We present a method for combining several segmentations of an image into a single one that in some sense is the average segmentation in order to achieve a more reliable and accurate segmentation