ML Anthology

An Improved Local Descriptor and Threshold Learning for Unsupervised Dynamic Texture Segmentation

Jie Chen, Guoying Zhao, Matti Pietikäinen
ICCVW 2009 pp. 460-467
doi:10.1109/ICCVW.2009.5457664 /iccvw/2009/chen2009iccvw-improved/

Abstract

Dynamic texture (DT) is an extension of texture to the temporal domain. How to segment DTs is a challenging problem. In this paper, we propose significant improvements to a recently published DT segmentation method. We employ a new spatiotemporal local texture descriptor which combines local binary patterns with a differential excitation measure. We also address the important problem of threshold selection by proposing a method for determining thresholds for the segmentation method by statistical learning. An improved criterion for merging adjacent regions is also introduced. Experimental results show that our approach provides very good segmentation results compared to state-of-the-art methods.

ICCVW Semantic Scholar

Cite

Text

Chen et al. "An Improved Local Descriptor and Threshold Learning for Unsupervised Dynamic Texture Segmentation." IEEE/CVF International Conference on Computer Vision Workshops, 2009. doi:10.1109/ICCVW.2009.5457664

Markdown

[Chen et al. "An Improved Local Descriptor and Threshold Learning for Unsupervised Dynamic Texture Segmentation." IEEE/CVF International Conference on Computer Vision Workshops, 2009.](https://mlanthology.org/iccvw/2009/chen2009iccvw-improved/) doi:10.1109/ICCVW.2009.5457664

BibTeX

@inproceedings{chen2009iccvw-improved,
  title     = {{An Improved Local Descriptor and Threshold Learning for Unsupervised Dynamic Texture Segmentation}},
  author    = {Chen, Jie and Zhao, Guoying and Pietikäinen, Matti},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2009},
  pages     = {460-467},
  doi       = {10.1109/ICCVW.2009.5457664},
  url       = {https://mlanthology.org/iccvw/2009/chen2009iccvw-improved/}
}