Robustizing robust M-estimation using deterministic annealing

This paper presents a modified robust M-estimator referred to as the annealing M-estimator (AM-estimator) to avoid problems with the M-estimator. The AM-estimator combines the annealing technique into the M-estimator. It has the following advantages: it approximates the global solution regardless of the initialization. It involves no scale estimator nor free parameters, avoiding the unreliability therein, nor does it need order statistics such as the median and hence no sorting. Experimental results show that the AM-estimator is very stable and has an elegant behavior with respect to percentage of outliers and noise variance.