Thanks Christoph for your reply.
Since the library does not support this specific case I am considering to implement by myself this algorithm.
It seems simple and efficient
Since the library does not support this specific case I am considering to implement by myself this algorithm.
It seems simple and efficient