CGAL users discussion list

[Marc Pouget <>]

Hello,

i summarize here the discussion we had by phone:

you can use interpolation with the jet fitting if you give exactly  
the nb of points that correspond to the nb of constraints.

Then, by default, the differential quantities returned are those  
computed at the projection of the first point you have given on the  
fitted polynomial surface.

If you want to compute the quantities for another point p, and you  
dont want this point to be taken into account to do the fitting  
(either-wise just give this point first and you are done); you can  
just project this point on the fitted surface (that is given by the  
monge form in the monge coordinate system) and compute the diff   
quantities by classical diff geo formula.

I may consider adding this functionality in the future ...

best
Marc


On 28 , May , 2008, at 06:42 , Olumide wrote:

> Hello -
>
> I'm looking to use the jet-fitting algorithm to compute the  
> principal curvature and directions at various non-vertex points of  
> a (triangular mesh) i.e. my points of interest lie on the faces of  
> a mesh, and by definition, do not coincide with its vertexes.
>
> My question however concerns the method used by the implementation  
> of the jet-fitting algorithm in this case. I would like to confirm  
> if the jet-fitting algorithm uses an (a) interpolation or (b)  
> approximation approach to compute the differential properties at  
> such no-vertex points -- the former would be preferred.
>
> I'm aware that this issue was discussed on page 12 of "Jet fitting  
> 3: A Generic C++ Package for Estimating the Differential Properties  
> on Sampled Surfaces via Polynomial Fitting", but I found the  
> reading a bit confusing probably because I'm only just learning the  
> algorithm.
>
> Thanks,
>
> - Olumide
>
>
> --
> You are currently subscribed to cgal-discuss.
> To unsubscribe or access the archives, go to
> https://lists-sop.inria.fr/wws/info/cgal-discuss