Dense Shape Correspondences Using Spectral High-Order Graph Matching

Abstract

This paper addresses the problem of establishing point correspondences between two object instances using spectral high-order graph matching. Therefore, 3D objects are intrinsically represented by weighted high-order adjacency tensors. These are, depending on the weighting scheme, invariant for structure-preserving, equi-areal, conformal or volume-preserving object deformations. Higher-order spectral decomposition transforms the NP-hard assignment problem into a linear assignment problem by canonical embedding. This allows to extract dense correspondence information with reasonable computational complexity, making the method faster than any other previously published method imposing higher-order constraints to shape matching. Robustness against missing data and resampling is measured and compared with a baseline spectral graph matching method. © 2011 IEEE.

Cite

Text

Smeets et al. "Dense Shape Correspondences Using Spectral High-Order Graph Matching." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2011. doi:10.1109/CVPRW.2011.5981675

Markdown

[Smeets et al. "Dense Shape Correspondences Using Spectral High-Order Graph Matching." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2011.](https://mlanthology.org/cvprw/2011/smeets2011cvprw-dense/) doi:10.1109/CVPRW.2011.5981675

BibTeX

@inproceedings{smeets2011cvprw-dense,
  title     = {{Dense Shape Correspondences Using Spectral High-Order Graph Matching}},
  author    = {Smeets, Dirk and Hermans, Jeroen and Vandermeulen, Dirk and Suetens, Paul},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2011},
  pages     = {1-8},
  doi       = {10.1109/CVPRW.2011.5981675},
  url       = {https://mlanthology.org/cvprw/2011/smeets2011cvprw-dense/}
}