CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Tractable Algorithms for Robust Model Estimation

Olof Enqvist (Institutionen för signaler och system, Bildanalys och datorseende) ; Erik Ask ; Fredrik Kahl (Institutionen för signaler och system, Bildanalys och datorseende) ; Kalle Åström
International Journal of Computer Vision (0920-5691). Vol. 112 (2015), 1, p. 115-129 .
[Artikel, refereegranskad vetenskaplig]

What is the computational complexity of geometric model estimation in the presence of noise and outliers? We show that the number of outliers can be minimized in polynomial time with respect to the number of measurements, although exponential in the model dimension. Moreover,for a large class of problems, we prove that the statistically more desirable truncated L2-norm can be optimized with the same complexity. In a similar vein, it is also shown how to transform a multi-model estimation problem into a purely combinatorial one—with worst-case complexity that is polynomial in the number of measurements but exponential in the number of models. We apply our framework to a series of hard fitting problems. It gives a practical method for simultaneously dealing with measurement noise and large amounts of outliers in the estimation of low-dimensional models. Experimental results and a comparison to random sampling techniques are presented for the applications rigid registration, triangulation and stitching.



Denna post skapades 2014-10-16. Senast ändrad 2016-05-24.
CPL Pubid: 204354

 

Läs direkt!


Länk till annan sajt (kan kräva inloggning)