Point Pattern Matching with Robust Spectral Correspondence

Abstract

This paper investigates the correspondence matching of point-sets using spectral graph analysis. In particular we are interested in the problem of how the modal analysis of point-sets can be rendered robust to contamination and drop-out. We make three contributions. First, we show how the modal structure of point-sets can be embedded within the framework of the EM algorithm. Second, we present several methods for computing the probabilities of point correspondences using the point proximity matrix. Third, we consider alternatives to the Gaussian proximity matrix. We evaluate the new method on both synthetic and real-world data. Here we show that the method can be used to compute useful correspondences even when the level of point contamination is as large as 50%.

Cite

Text

Carcassoni and Hancock. "Point Pattern Matching with Robust Spectral Correspondence." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2000. doi:10.1109/CVPR.2000.855881

Markdown

[Carcassoni and Hancock. "Point Pattern Matching with Robust Spectral Correspondence." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2000.](https://mlanthology.org/cvpr/2000/carcassoni2000cvpr-point/) doi:10.1109/CVPR.2000.855881

BibTeX

@inproceedings{carcassoni2000cvpr-point,
  title     = {{Point Pattern Matching with Robust Spectral Correspondence}},
  author    = {Carcassoni, Marco and Hancock, Edwin R.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2000},
  pages     = {1649-1655},
  doi       = {10.1109/CVPR.2000.855881},
  url       = {https://mlanthology.org/cvpr/2000/carcassoni2000cvpr-point/}
}