Authors:
Jean-Philippe Tarel
1
;
Pierre Charbonnier
2
and
Sio-Song Ieng
3
Affiliations:
1
ESE, Laboratoire Central des Ponts et Chaussées, France
;
2
ERA 27 LCPC, Laboratoire des Ponts et Chaussées, France
;
3
ERA 17 LCPC, Laboratoire des Ponts et Chaussées, France
Keyword(s):
Image Analysis, Statistical Approach, Robust Fitting, Multiple Fitting, Image Grouping and Segmentation.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Image and Video Analysis
;
Segmentation and Grouping
;
Statistical Approach
Abstract:
In this paper, we address the problem of robustly recovering several instances of a curve model from a single noisy data set with outliers. Using M-estimators revisited in a Lagrangian formalism, we derive an algorithm that we call SMRF, which extends the classical Iterative Reweighted Least Squares algorithm (IRLS). Compared to the IRLS, it features an extra probability ratio, which is classical in clustering algorithms, in the expression of the weights. Potential numerical issues are tackled by banning zero probabilities in the computation of the weights and by introducing a Gaussian prior on curves coefficients. Applications to camera calibration and lane-markings tracking show the effectiveness of the SMRF algorithm, which outperforms classical Gaussian mixture model algorithms in the presence of outliers.