CGAL users discussion list

["Oliver Arend" <>]

I have a file with a bunch of 3D points stored in binary. Now I want to both
estimate the normal and then use these to do a Poisson surface
reconstruction. Unfortunately the example for normal estimation at 
http://doc.cgal.org/latest/Point_set_processing_3/Point_set_processing_3_2no
rmals_example_8cpp-example.html
uses

typedef std::pair<Point, Vector> PointVectorPair;
std::list<PointVectorPair> points;

to store the points, and accesses points and normals through

CGAL::First_of_pair_property_map<PointVectorPair>
CGAL::Second_of_pair_property_map<PointVectorPair>

respectively.

After I have successfully estimated the normals for all my points I would
like to reconstruct the surface using Poisson. The example at
http://doc.cgal.org/latest/Poisson_surface_reconstruction_3/Poisson_surface_
reconstruction_3_2poisson_reconstruction_example_8cpp-example.html
uses

typedef CGAL::Point_with_normal_3<Kernel> Point_with_normal;
typedef std::vector<Point_with_normal> PointList;
PointList points;

to store the points.

Now it seems I can't simply pass the <Point, Vector> pair like in the
example

Poisson_reconstruction_function function(points.begin(), points.end(),
CGAL::make_normal_of_point_with_normal_pmap(PointList::value_type()) );

(even if I define PointList to be std::list<PointVectorPair>)

Please excuse my ignorance, but I am at a complete loss as to how to
approach this.

So how can I either
- do the whole normal estimation and Poisson reconstruction using a single
set of points (one of the above) or
- convert from <Point, Vector> pairs to Point_with_normal_3?

Thanks in advance
Oliver