ML Anthology

Do We Really Have to Consider Covariance Matrices for Image Features?

Yasushi Kanazawa, Ken-ichi Kanatani
ICCV 2001 pp. 301-306
doi:10.1109/ICCV.2001.937640 /iccv/2001/kanazawa2001iccv-we/

Abstract

Many studies have been made in the past for optimization using covariance matrices of feature points. We first describe how to compute the covariance matrix of a feature point from the gray levels by integrating existing methods. Then, we experimentally examine if thus computed covariance matrices really reflect the accuracy of the feature points. To test this, we do subpixel template matching and compute the homography and the fundamental matrix. Our conclusion is rather surprising, pointing out important elements often overlooked.

ICCV Semantic Scholar

Cite

Text

Kanazawa and Kanatani. "Do We Really Have to Consider Covariance Matrices for Image Features?." IEEE/CVF International Conference on Computer Vision, 2001. doi:10.1109/ICCV.2001.937640

Markdown

[Kanazawa and Kanatani. "Do We Really Have to Consider Covariance Matrices for Image Features?." IEEE/CVF International Conference on Computer Vision, 2001.](https://mlanthology.org/iccv/2001/kanazawa2001iccv-we/) doi:10.1109/ICCV.2001.937640

BibTeX

@inproceedings{kanazawa2001iccv-we,
  title     = {{Do We Really Have to Consider Covariance Matrices for Image Features?}},
  author    = {Kanazawa, Yasushi and Kanatani, Ken-ichi},
  booktitle = {IEEE/CVF International Conference on Computer Vision},
  year      = {2001},
  pages     = {301-306},
  doi       = {10.1109/ICCV.2001.937640},
  url       = {https://mlanthology.org/iccv/2001/kanazawa2001iccv-we/}
}