| http://www.w3.org/ns/prov#value | - The TV norm is basically the l^1 norm of the (discretised) gradient nabla f of f, and so one can deduce exact recovery of TV norm for signals of sparse gradient from the general theory of exact recovery of l^1 norms for sparse signals [one has to extend this theory to vector-valued signals, since the gradient is a vector, but this is routine]. (Note that the gradient is not free to range over all
|
