Abstract
We propose a solution to the problem of robust subspace es- timation using the pro jection based M-estimator. The new method han- dles more outliers than inliers, does not require a user defined scale of the noise affecting the inliers, handles noncentered data and nonorthog- onal subspaces. Other robust methods like RANSAC, use an input for the scale, while methods for subspace segmentation, like GPCA, are not robust. Synthetic data and three real cases of multibody factorization show the superiority of our method, in spite of user independence.