Free Online Productivity Tools
i2Speak
i2Symbol
i2OCR
iTex2Img
iWeb2Print
iWeb2Shot
i2Type
iPdf2Split
iPdf2Merge
i2Bopomofo
i2Arabic
i2Style
i2Image
i2PDF
iLatex2Rtf
Sci2ools

PRL

2010

2010

We consider the problem of ﬁtting linearly parameterized models, that arises in many computer vision problems such as road scene analysis. Data extracted from images usually contain non-Gaussian noise and outliers, which makes non-robust estimation methods ineﬀective. In this paper, we propose an overview of a Lagrangian formulation of the Half-Quadratic approach by, ﬁrst, revisiting the derivation of the well-known Iterative Re-weighted Least Squares (IRLS) robust estimation algorithm. Then, it is shown that this formulation helps derive the so-called Modiﬁed Residuals Least Squares (MRLS) algorithm. In this framework, moreover, standard theoretical results from constrained optimization can be invoked to derive convergence proofs easier. The interest of using the Lagrangian framework is also illustrated by the extension to the problem of the robust estimation of sets of linearly parameterized curves, and to the problem of robust ﬁtting of linearly parameterized regions. To d...

Added |
30 Jan 2011 |

Updated |
17 Feb 2011 |

Type |
Journal |

Year |
2010 |

Where |
PRL |

Authors |
Jean-Philippe Tarel, Pierre Charbonnier |

Reference is available at

http://perso.lcpc.fr/tarel.jean-philippe/publis/prl10.html

http://perso.lcpc.fr/tarel.jean-philippe/publis/prl10.html

Comments (0)