Shape classification of partially occluded objects using subspace detectors (TP-P4)

Author(s) : Arnt B. Salberg (Institute of Marine Research, Dept. Tromsų, Norway) Alf Harbitz (Institute of Marine Research, Dept. Tromsų, Norway) Alfred Hanssen (Univsersity of Tromsų, Dept. Physics, Norway)

Abstract : In this paper we consider shape classification of partially occluded objects. We show that robust subspace detectors can be formulated as a weighted subspace detector, and the elements in the boundary vector will be weighted before they are matched. The part of the boundary vector that orresponds to the occluded part of the object, will be suppressed by the weight vector, and hence have a reduced effect on the classification performance.

Menu