Semantic 3D Reconstruction with Finite Element Bases

Audrey Richard, Christoph Vogel, Maros Blaha, Thomas Pock and Konrad Schindler

Abstract

We propose a novel framework for the discretisation of multi-label problems on arbitrary, continuous domains. Our work bridges the gap between general FEM discretisations, and labeling problems that arise in a variety of computer vision tasks, including for instance those derived from the generalised Potts model. Starting from the popular formulation of labeling as a convex relaxation by functional lifting, we show that FEM discretisation is valid for the most general case, where the regulariser is anisotropic and non-metric. While our findings are generic and applicable to different vision problems, we demonstrate their practical implementation in the context of semantic 3D reconstruction, where such regularisers have proved particularly beneficial.

Session

Spotlights

Files

Paper (PDF)
Supplementary (PDF)
Video (MP4)

DOI

10.5244/C.31.98
https://dx.doi.org/10.5244/C.31.98

Citation

Bibtex

            @inproceedings{BMVC2017_98,
                title={Semantic 3D Reconstruction with Finite Element Bases},
                author={Audrey Richard, Christoph Vogel, Maros Blaha, Thomas Pock and Konrad Schindler},
                year={2017},
                month={September},
                pages={98.1-98.13},
                articleno={98},
                numpages={13},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.98},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.98}
            }