Triadic Split-Merge Sampler

Anne C. van Rossum, Hai Xiang Lin, Johan Dubbeldam, Jaap van den Herik

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review


In machine vision typical heuristic methods to extract parameterized objects out of raw data points are the Hough transform and RANSAC. Bayesian models carry the promise to optimally extract such parameterized objects given a correct definition of the model and the type of noise at hand. A category of solvers for Bayesian models are Markov chain Monte Carlo methods. Naive implementations of MCMC methods suffer from slow convergence in machine vision due to the complexity of the parameter space. Towards this blocked Gibbs and split-merge samplers have been developed that assign multiple data points to clusters at once. In this paper we introduce a new split-merge sampler, the triadic split-merge sampler, that perform steps between two and three randomly chosen clusters. This has two advantages. First, it reduces the asymmetry between the split and merge steps. Second, it is able to propose a new cluster that is composed out of data points from two different clusters. Both advantages speed up convergence which we demonstrate on a line extraction problem. We show that the triadic split-merge sampler outperforms the conventional split-merge sampler. Although this new MCMC sampler is demonstrated in this machine vision context, its application extend to the very general domain of statistical inference.
Original languageEnglish
Title of host publicationProceedings of SPIE: Tenth International Conference on Machine Vision (ICMV 2017)
EditorsAntanas Verikas, Petia Radeva, Dmitry Nikolaev, Jianhong Zhou
Number of pages8
ISBN (Print)9781510619418
Publication statusPublished - Apr 2018
Event10th International Conference on Machine Vision - Vienna, Austria
Duration: 13 Nov 201715 Nov 2017
Conference number: 10


Conference10th International Conference on Machine Vision
Abbreviated title(ICMV 2017


  • Bayesian statistics
  • MCMC
  • Split-Merge Sampler


Dive into the research topics of 'Triadic Split-Merge Sampler'. Together they form a unique fingerprint.

Cite this