Grouping with bias for distribution-free mixture model estimation

Abstract

Some authors have recently devised adaptations of spectral grouping algorithms to integrate prior knowledge, as constrained eigenvalues problems. In this paper, we adapt recent statistical grouping algorithms to this task, as a nonparametric mixture model estimation problem. The approach appears to be attractive for its theoretical benefits, and its experimental results, as light bias brings dramatic improvements over unbiased approaches on hard images. Show more