Efficient Minimal-Surface Regularization of Perspective Depth Maps in Variational Stereo

Gottfried Graber, Jonathan Balzer, Stefano Soatto, Thomas Pock

CVPR 2015

doi:10.1109/CVPR.2015.7298649 /cvpr/2015/graber2015cvpr-efficient/

Abstract

We propose a method for dense three-dimensional surface reconstruction that leverages the strengths of shape-based approaches, by imposing regularization that respects the geometry of the surface, and the strength of depth-map-based stereo, by avoiding costly computation of surface topology. The result is a near real-time variational reconstruction algorithm free of the staircasing artifacts that affect depth-map and plane-sweeping approaches. This is made possible by exploiting the gauge ambiguity to design a novel representation of the regularizer that is linear in the parameters and hence amenable to be optimized with state-of-the-art primal-dual numerical schemes.

PDF CVPR Semantic Scholar

Cite

Text

Graber et al. "Efficient Minimal-Surface Regularization of Perspective Depth Maps in Variational Stereo." Conference on Computer Vision and Pattern Recognition, 2015. doi:10.1109/CVPR.2015.7298649

Markdown

[Graber et al. "Efficient Minimal-Surface Regularization of Perspective Depth Maps in Variational Stereo." Conference on Computer Vision and Pattern Recognition, 2015.](https://mlanthology.org/cvpr/2015/graber2015cvpr-efficient/) doi:10.1109/CVPR.2015.7298649

BibTeX

@inproceedings{graber2015cvpr-efficient,
  title     = {{Efficient Minimal-Surface Regularization of Perspective Depth Maps in Variational Stereo}},
  author    = {Graber, Gottfried and Balzer, Jonathan and Soatto, Stefano and Pock, Thomas},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2015},
  doi       = {10.1109/CVPR.2015.7298649},
  url       = {https://mlanthology.org/cvpr/2015/graber2015cvpr-efficient/}
}