Detection of point landmarks in multidimensional tensor data

This paper describes a unified approach to the detection of point landmarks -whose neighborhoods convey discriminant information-including multidimensi onal scalar, vector, and higher-order tensor data. The method is based on t he interpretation of generalized correlation matrices derived from the grad ient of tensor functions, a probabilistic interpretation of point landmarks , and the application of tensor algebra. Results on both synthetic and real tensor data are presented. (C) 2001 Elsevier Science B.V. All rights reser ved.