Skip to Content.
Sympa Menu

cgal-discuss - [cgal-discuss] CGAL newbie question

Subject: CGAL users discussion list

List archive

[cgal-discuss] CGAL newbie question


Chronological Thread 
  • From: Manu Kaul <>
  • To:
  • Subject: [cgal-discuss] CGAL newbie question
  • Date: Sun, 9 Sep 2012 08:18:29 +0200

Hi all,
I used the "Triangle" library from http://www.cs.cmu.edu/~quake/triangle.html by Jonathan Shewchuck to generate a Delaunay triangulation 
of my raw x,y,z points. I am trying to create an indexing structure and some distance bounds that make the shortest path calculation on a 
terrain faster + some other queries like range queries, and k-nearest neighbour queries as well. 

I noticed that in order to do some of these things I might need CGAL, but I understand that the learning curve can be quite steep. 
The things I need to do are :
1. Generate TINs at multiple levels of details by introducing Steiner points. 
2. Being able to access the triangles and its adjacent triangles in raw form (like 0 1 2 vertices of triangle in counter-clockwise format)
so that I can feed it to a "Chen and Han" shortest path implementation by O'Rourke (http://cs.smith.edu/~orourke/ShortestPaths/
3. Have the ability to compute the slope of each triangle and also all its interior angles.
4. Need to be able to *project* onto the terrain the 3D euclidean line between any two points of the terrain.
5. Generate a Voronoi diagram from the Delaunay (Triangle generates this too)

Someone who used CGAL mentioned that traversing over faces of a triangulation in CGAL can be quite cumbersome and that I should
instead convert it to a polyhedron and use the iterator there. So I just want to understand if CGAL will be able to help me with all the 
things I need to do and if I might get stuck halfway? Hopefully CGAL will not be overkill for what I am trying to do? I found another C++
library that used the DECL (http://www.holmes3d.net/graphics/dcel/), would this be easier to use than CGAL for what I am trying to achieve?
I ask since I am not going to do very complicated computational geometry related work on the surface, if CGAL would be too much to work with?

Thanks,
Manu

--

The greater danger for most of us lies not in setting our aim too high and falling short; but in setting our aim too low, and achieving our mark.
- Michelangelo



Archive powered by MHonArc 2.6.18.

Top of Page