Header

UZH-Logo

Maintenance Infos

SOAR: Stochastic optimization for affine global point set registration


Agus, Marco; Gobbetti, Enrico; Villanueva, Alberto Jaspe; Mura, Claudio; Pajarola, R (2014). SOAR: Stochastic optimization for affine global point set registration. In: Proceedings Vision, Modeling and Visualization, Darmstadt, Germany, 8 October 2014 - 10 October 2014, 103-110.

Abstract

We introduce a stochastic algorithm for pairwise affine registration of partially overlapping 3D point clouds with unknown point correspondences. The algorithm recovers the globally optimal scale, rotation, and translation alignment parameters and is applicable in a variety of difficult settings, including very sparse, noisy, and outlier-ridden datasets that do not permit the computation of local descriptors. The technique is based on a stochastic approach for the global optimization of an alignment error function robust to noise and resistant to outliers. At each optimization step, it alternates between stochastically visiting a generalized BSP-tree representation of the current solution landscape to select a promising transformation, finding point-to-point correspondences using a GPU-accelerated technique, and incorporating new error values in the BSP tree. In contrast to previous work, instead of simply constructing the tree by guided random sampling, we exploit the problem structure through a low-cost local minimization process based on analytically solving absolute orientation problems using the current correspondences. We demonstrate the quality and performance of our method on a variety of large point sets with different scales, resolutions, and noise characteristics.

Abstract

We introduce a stochastic algorithm for pairwise affine registration of partially overlapping 3D point clouds with unknown point correspondences. The algorithm recovers the globally optimal scale, rotation, and translation alignment parameters and is applicable in a variety of difficult settings, including very sparse, noisy, and outlier-ridden datasets that do not permit the computation of local descriptors. The technique is based on a stochastic approach for the global optimization of an alignment error function robust to noise and resistant to outliers. At each optimization step, it alternates between stochastically visiting a generalized BSP-tree representation of the current solution landscape to select a promising transformation, finding point-to-point correspondences using a GPU-accelerated technique, and incorporating new error values in the BSP tree. In contrast to previous work, instead of simply constructing the tree by guided random sampling, we exploit the problem structure through a low-cost local minimization process based on analytically solving absolute orientation problems using the current correspondences. We demonstrate the quality and performance of our method on a variety of large point sets with different scales, resolutions, and noise characteristics.

Statistics

Citations

Altmetrics

Downloads

18 downloads since deposited on 22 Jan 2015
14 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Conference or Workshop Item (Paper), refereed, original work
Communities & Collections:03 Faculty of Economics > Department of Informatics
Dewey Decimal Classification:000 Computer science, knowledge & systems
Uncontrolled Keywords:graphics, laser scanning, point cloud, registration
Language:English
Event End Date:10 October 2014
Deposited On:22 Jan 2015 15:12
Last Modified:15 Aug 2017 21:48
Publisher:The Eurographics Association
Publisher DOI:https://doi.org/10.2312/vmv.20141282
Official URL:http://www.vmv2014.de/
Other Identification Number:merlin-id:10256

Download

Download PDF  'SOAR: Stochastic optimization for affine global point set registration'.
Preview
Filetype: PDF
Size: 610kB
View at publisher