Multi-H: Efficient recovery of tangent planes in stereo images

Daniel Barath, Jiri Matas and Levente Hajder

Abstract

Multi-H _ an efficient method for the recovery of the tangent planes of a set of point correspondences satisfying the epipolar constraint is proposed. The problem is formulated as a search for a labeling minimizing an energy that includes a data and spatial regularization terms. The number of planes is controlled by a combination of Mean- Shift and $\alpha$-expansion. Experiments on the fountain-P11 3D dataset show that Multi-H provides highly accurate tangent plane estimates. It also outperforms all state-of-the-art techniques for multihomography estimation on the publicly available AdelaideRMF dataset. Since Multi-H achieves nearly error-free performance, we introduce and make public a more challenging dataset for multi-plane fitting evaluation.

Session

Posters 1

Files

PDF iconExtended Abstract (PDF, 4M)
PDF iconPaper (PDF, 19M)

DOI

10.5244/C.30.13
https://dx.doi.org/10.5244/C.30.13

Citation

Daniel Barath, Jiri Matas and Levente Hajder. Multi-H: Efficient recovery of tangent planes in stereo images. In Richard C. Wilson, Edwin R. Hancock and William A. P. Smith, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 13.1-13.13. BMVA Press, September 2016.

Bibtex

        @inproceedings{BMVC2016_13,
        	title={Multi-H: Efficient recovery of tangent planes in stereo images},
        	author={Daniel Barath, Jiri Matas and Levente Hajder},
        	year={2016},
        	month={September},
        	pages={13.1-13.13},
        	articleno={13},
        	numpages={13},
        	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
        	publisher={BMVA Press},
        	editor={Richard C. Wilson, Edwin R. Hancock and William A. P. Smith},
        	doi={10.5244/C.30.13},
        	isbn={1-901725-59-6},
        	url={https://dx.doi.org/10.5244/C.30.13}
        }