Improving Bayesian Mixture Models for Colour Image Segmentation with Superpixels

Thorsten Wilhelm, Christian Wöhler

2017

Abstract

The large computational demand is one huge drawback of Bayesian Mixture Models in image segmentation tasks. We describe a novel approach to reduce the computational demand in this scenario and increase the performance by using superpixels. Superpixels provide a natural approach to the reduction of the computational complexity and to build a texture model in the image domain. Instead of relying on a Gaussian mixture model as segmentation model, we propose to use a more robust model: a mixture of multiple scaled t-distributions. The parameters of the novel mixture model are estimated with Markov chain Monte Carlo in order to surpass local minima during estimation and to gain insight into the uncertainty of the resulting segmentation. Finally, an evaluation of the proposed segmentation is performed on the publicly available Berkeley Segmentation database (BSD500), compared to competing methods, and the benefit of including texture is emphasised.

Paper Citation

in Harvard Style

Wilhelm T. and Wöhler C. (2017). Improving Bayesian Mixture Models for Colour Image Segmentation with Superpixels . In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2017) ISBN 978-989-758-225-7, pages 443-450. DOI: 10.5220/0006111504430450

in Bibtex Style

@conference{visapp17,
author={Thorsten Wilhelm and Christian Wöhler},
title={Improving Bayesian Mixture Models for Colour Image Segmentation with Superpixels},
booktitle={Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2017)},
year={2017},
pages={443-450},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006111504430450},
isbn={978-989-758-225-7},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2017)
TI - Improving Bayesian Mixture Models for Colour Image Segmentation with Superpixels
SN - 978-989-758-225-7
AU - Wilhelm T.
AU - Wöhler C.
PY - 2017
SP - 443
EP - 450
DO - 10.5220/0006111504430450