Robust Multiview Registration of 3D Surfaces via L_1-norm Minimization

Anil C. Raghuramu

Abstract

In this paper we present a robust method for simultaneous registration of multiple 3D scans. Rigid registration is an important task in many applications such as surface reconstruction, navigation and computer aided design. The goal of 3-D (rigid) registration is to align surfaces through a (rigid) transformation. A large number of existing registration algorithms are dependent on finding matching points between scans, but a significant number of them are spurious, and it is necessary to clean up the matches obtained. This requires a substantial amount of tuning of parameters and the final result might still contain outliers. Since the number of outliers are sparse we formulate the registration optimization using the $\ell_1$-norm. We present experimental results to show that the performance of our algorithm is comparable to state of the art algorithms.

Session

Poster 1

Files

PDF iconExtended Abstract (PDF, 407K)
PDF iconPaper (PDF, 4M)

DOI

10.5244/C.29.62
https://dx.doi.org/10.5244/C.29.62

Citation

Anil C. Raghuramu. Robust Multiview Registration of 3D Surfaces via L_1-norm Minimization. In Xianghua Xie, Mark W. Jones, and Gary K. L. Tam, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 62.1-62.11. BMVA Press, September 2015.

Bibtex

@inproceedings{BMVC2015_62,
	title={Robust Multiview Registration of 3D Surfaces via L_1-norm Minimization},
	author={Anil C. Raghuramu},
	year={2015},
	month={September},
	pages={62.1-62.11},
	articleno={62},
	numpages={11},
	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
	publisher={BMVA Press},
	editor={Xianghua Xie, Mark W. Jones, and Gary K. L. Tam},
	doi={10.5244/C.29.62},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.62}
}