@inproceedings{7c9681ba299f4ef9bbc869c588049d08,
title = "Robust and Accurate Cylinder Triangulation",
abstract = "In this paper we present methods for triangulation of infinite cylinders from image line silhouettes. We show numerically that linear estimation of a general quadric surface is inherently a badly posed problem. Instead we propose to constrain the conic section to a circle, and give algebraic constraints on the dual conic, that models this manifold. Using these constraints we derive a fast minimal solver based on three image silhouette lines, that can be used to bootstrap robust estimation schemes such as RANSAC. We also present a constrained least squares solver that can incorporate all available image lines for accurate estimation. The algorithms are tested on both synthetic and real data, where they are shown to give accurate results, compared to previous methods.",
keywords = "Cylinders, Reconstruction, Robust estimation",
author = "Anna Gummeson and Magnus Oskarsson",
year = "2023",
doi = "10.1007/978-3-031-31438-4_30",
language = "English",
isbn = "9783031314377",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media B.V.",
pages = "451--466",
editor = "Rikke Gade and Michael Felsberg and Joni-Kristian K{\"a}m{\"a}r{\"a}inen",
booktitle = "Image Analysis - 23rd Scandinavian Conference, SCIA 2023, Proceedings",
address = "United States",
note = "23nd Scandinavian Conference on Image Analysis, SCIA 2023 ; Conference date: 18-04-2023 Through 21-04-2023",
}