D. Roobaert (1999)
Improving the Generalisation of Linear Support Vector Machines: an Application to 3D Object Recognition with Cluttered Background
In: Proc. SVM Workshop at IJCAI'99, Stockholm, Sweden.
Three methods for improving the generalisation of linear SVMs are proposed in the case some dimensions in the data can be considered irrelevant for the pattern recognition task at hand. As application 3D object recognition with cluttered background is performed, for which very good results are reported using the black/white background training method.