I try to use inverse matrix procedure of vnl but I've some trouble to compile :
inv = vnl_matrix_inverse(A);
use of class template `template<class T> struct vnl_matrix_inverse' as
What's wrong ?
If anybody can help me,
On Thu 15 Apr 2004, Rousseau Francois wrote:
> vnl_matrix<double> A(4,4);
> vnl_matrix<double> inv(4,4);
> inv = vnl_matrix_inverse(A);
I'm afraid the documentation of that class is slightly incorrect. Try
inv = vnl_matrix_inverse<double>(A);
Also: if you matrix is 4x4 or smaller, you may want to look at
vnl_double_4x4 (a convenient interface to
vnl_matrix_fixed<double,4,4>) and vnl_inverse which does the inverse
directly (and is thus much faster than computing the SVD).