By the same authors

From the same journal

From the same journal

Probabilistic relaxation labelling using the Fokker-Planck equation

Research output: Contribution to journalArticle

Published copy (DOI)



Publication details

JournalPattern recognition
DatePublished - Nov 2008
Issue number11
Number of pages19
Pages (from-to)3393-3411
Original languageEnglish


In this paper we develop a new formulation of probabilistic relaxation labelling using the theory of diffusion processes on graphs. Our aim is to tackle the problem of labelling objects consistently and unambiguously using information concerning label consistency and initial label probabilities. We abstract this problem using a support graph with each graph node an object-label assignment. Initial object-label probabilities then evolve across the graph under the governance of the Fokker-Planck equation in terms of an infinitesimal generator matrix computed from the edge weights of the Support graph. In this way we effectively kernelise probabilistic relaxation. Encouraging results are obtained in applying the new relaxation process in the applications of scene labelling, data classification, and feature correspondence matching. (C) 2008 Elsevier Ltd. All rights reserved.

    Research areas

  • data clustering, feature correspondence matching, scene labelling, relaxation labelling, graph theory, diffusion process, Fokker-Planck equation, DISCRETE RELAXATION, ALGORITHM, MATRIX, IMAGES

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations