Generalized weiszfeld algorithms for lq optimization
In many computer vision applications, a desired model of some type is computed by minimizing a cost function based on several measurements. Typically, one may compute the model that minimizes the L2 cost, that is the sum of squares of measurement errors with respect to the model. The simplicity of L2 methods is well appreciated, but they suffer from a drawback when it comes to robustness against outliers; even a single outlier may change the L2 solution drastically. This thesis is primarily...[Show more]
|Collections||Open Access Theses|
|b37326867_Aftab_Khurrum.pdf||276.55 MB||Adobe PDF|
Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.