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 depthmap-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.
the geometry of the surface, and the strength of depthmap-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.
Originalsprache | englisch |
---|---|
Titel | The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) |
Herausgeber (Verlag) | . |
Seiten | 511-520 |
Publikationsstatus | Veröffentlicht - 2015 |
Veranstaltung | 2015 IEEE Conference on Computer Vision and Pattern Recognition: CVPR 2015 - Boston, USA / Vereinigte Staaten Dauer: 7 Juni 2015 → 12 Juni 2015 |
Konferenz
Konferenz | 2015 IEEE Conference on Computer Vision and Pattern Recognition |
---|---|
Kurztitel | CVPR 2015 |
Land/Gebiet | USA / Vereinigte Staaten |
Ort | Boston |
Zeitraum | 7/06/15 → 12/06/15 |
Fields of Expertise
- Information, Communication & Computing
Treatment code (Nähere Zuordnung)
- Application