ML Anthology

A New Semi-Supervised EM Algorithm for Image Retrieval

Anlei Dong, Bir Bhanu
CVPR 2003 pp. 662-667
doi:10.1109/CVPR.2003.1211530 /cvpr/2003/dong2003cvpr-new/

Abstract

One of the main tasks in content-based image retrieval (CBIR) is to reduce the gap between low-level visual features and high-level human concepts. This paper presents a new semi-supervised EM algorithm (NSSEM), where the image distribution in feature space is modeled as a mixture of Gaussian densities. Due to the statistical mechanism of accumulating and processing meta knowledge, the NSS-EM algorithm with long term learning of mixture model parameters can deal with the cases where users may mislabel images during relevance feedback. Our approach that integrates mixture model of the data, relevance feedback and long term learning helps to improve retrieval performance. The concept learning is incrementally refined with increased retrieval experiences. Experiment results on Corel database show the efficacy of our proposed concept learning approach.

CVPR Semantic Scholar

Cite

Text

Dong and Bhanu. "A New Semi-Supervised EM Algorithm for Image Retrieval." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2003. doi:10.1109/CVPR.2003.1211530

Markdown

[Dong and Bhanu. "A New Semi-Supervised EM Algorithm for Image Retrieval." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2003.](https://mlanthology.org/cvpr/2003/dong2003cvpr-new/) doi:10.1109/CVPR.2003.1211530

BibTeX

@inproceedings{dong2003cvpr-new,
  title     = {{A New Semi-Supervised EM Algorithm for Image Retrieval}},
  author    = {Dong, Anlei and Bhanu, Bir},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2003},
  pages     = {662-667},
  doi       = {10.1109/CVPR.2003.1211530},
  url       = {https://mlanthology.org/cvpr/2003/dong2003cvpr-new/}
}