Title :
Generalized Projection-Based M-Estimator
Author :
Mittal, Sushil ; Anand, Saket ; Meer, Peter
Author_Institution :
Dept. of Stat., Columbia Univ., New York, NY, USA
Abstract :
We propose a novel robust estimation algorithm - the generalized projection-based M-estimator (gpbM), which does not require the user to specify any scale parameters. The algorithm is general and can handle heteroscedastic data with multiple linear constraints for single and multicarrier problems. The gpbM has three distinct stages - scale estimation, robust model estimation, and inlier/outlier dichotomy. In contrast, in its predecessor pbM, each model hypotheses was associated with a different scale estimate. For data containing multiple inlier structures with generally different noise covariances, the estimator iteratively determines one structure at a time. The model estimation can be further optimized by using Grassmann manifold theory. We present several homoscedastic and heteroscedastic synthetic and real-world computer vision problems with single and multiple carriers.
Keywords :
computer vision; estimation theory; Grassmann manifold theory; computer vision problems; generalized projection-based M-estimator; gpbM; heteroscedastic data; inlier-outlier dichotomy; linear constraints; noise covariances; robust model estimation algorithm; scale estimation; Computational modeling; Covariance matrix; Estimation; Noise measurement; Robust estimation; Robustness; Generalized projection-based M-estimator; RANSAC; heteroscedasticity; robust estimation;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
DOI :
10.1109/TPAMI.2012.52