SVMs using geometric algebra for 3D computer vision
Eduardo Jose Bayro-Corrochano, R. Vallejo · 2002
This paper shows the analysis of support multivector machines using the coordinate-free system of Clifford or geometric algebra. Real-, complex- and quaternion-valued neural networks are simple particular cases of the geometric algebra multidimensional neural networks and that they can be generated using support multivector machines. Particularly, the generation of RBF for neurocomputing in geometric algebra is easier using the SMVM, which allows to find the optimal parameters automatically. The use of SVM in the geometric algebra framework expands its sphere of applicability for multidimensional learning. As illustration we present the estimation of 3D rigid motion and 3D pose of rigid objects using visual information captured by a trinocular head.