ML Anthology

A Convex Formulation of Continuous Multi-Label Problems

Thomas Pock, Thomas Schoenemann, Gottfried Graber, Horst Bischof, Daniel Cremers
ECCV 2008 pp. 792-805
doi:10.1007/978-3-540-88690-7_59 /eccv/2008/pock2008eccv-convex/

Abstract

We propose a spatially continuous formulation of Ishikawa’s discrete multi-label problem. We show that the resulting non-convex variational problem can be reformulated as a convex variational problem via embedding in a higher dimensional space. This variational problem can be interpreted as a minimal surface problem in an anisotropic Riemannian space. In several stereo experiments we show that the proposed continuous formulation is superior to its discrete counterpart in terms of computing time, memory efficiency and metrication errors.

PDF ECCV Semantic Scholar

Cite

Text

Pock et al. "A Convex Formulation of Continuous Multi-Label Problems." European Conference on Computer Vision, 2008. doi:10.1007/978-3-540-88690-7_59

Markdown

[Pock et al. "A Convex Formulation of Continuous Multi-Label Problems." European Conference on Computer Vision, 2008.](https://mlanthology.org/eccv/2008/pock2008eccv-convex/) doi:10.1007/978-3-540-88690-7_59

BibTeX

@inproceedings{pock2008eccv-convex,
  title     = {{A Convex Formulation of Continuous Multi-Label Problems}},
  author    = {Pock, Thomas and Schoenemann, Thomas and Graber, Gottfried and Bischof, Horst and Cremers, Daniel},
  booktitle = {European Conference on Computer Vision},
  year      = {2008},
  pages     = {792-805},
  doi       = {10.1007/978-3-540-88690-7_59},
  url       = {https://mlanthology.org/eccv/2008/pock2008eccv-convex/}
}