Dense parameter fields from total least squares

TitleDense parameter fields from total least squares
Publication TypeConference Paper
Year of Publication2002
AuthorsSpies, H, Garbe, CS
EditorVan Gool, L
Conference NameProceedings of the 24th DAGM Symposium on Pattern Recognition

A method for the interpolation of parameter fields estimated by total least squares is presented. This is applied to the study of dynamic processes where the motion and further values such as divergence or brightness changes are parameterised in a partial differential equation. For the regularisation we introduce a constraint that restricts the solution only in the subspace determined by the total least squares procedure. The performance is illustrated on both synthetic and real test data.


Lecture Notes in Computer Science

Citation Keyspies2002b