Mixture Models and the Segmentation of Multimodal Textures

Manduchi, Roberto

doi:10.1109/CVPR.2000.855805

Mixture Models and the Segmentation of Multimodal Textures

Roberto Manduchi

CVPR 2000 pp. 1098-1104

doi:10.1109/CVPR.2000.855805 /cvpr/2000/manduchi2000cvpr-mixture/

Abstract

A problem with using mixture-of-Gaussian models for unsupervised texture segmentation is that a "multimodal" texture (such as can often be encountered in natural images) cannot be well represented by a single Gaussian cluster. We propose a divide-and-conquer method that groups together Gaussian clusters (estimated via Expectation Maximization) into homogeneous texture classes. This method allows to successfully segment even rather complex textures, as demonstrated by experimental tests on natural images.

CVPR Semantic Scholar

Cite

Text

Manduchi. "Mixture Models and the Segmentation of Multimodal Textures." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2000. doi:10.1109/CVPR.2000.855805

Markdown

[Manduchi. "Mixture Models and the Segmentation of Multimodal Textures." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2000.](https://mlanthology.org/cvpr/2000/manduchi2000cvpr-mixture/) doi:10.1109/CVPR.2000.855805

BibTeX

@inproceedings{manduchi2000cvpr-mixture,
  title     = {{Mixture Models and the Segmentation of Multimodal Textures}},
  author    = {Manduchi, Roberto},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2000},
  pages     = {1098-1104},
  doi       = {10.1109/CVPR.2000.855805},
  url       = {https://mlanthology.org/cvpr/2000/manduchi2000cvpr-mixture/}
}