3 messages
Open this post in threaded view
|

 Hi list, I have several question about Mesh_3. 1. I'm generating isosurface by implicit function. It requires that the center of the sphere be inside of the mesh. The question is, given an arbitrary mesh, how can I find such a point? 2. Due to the first problem(I guess), the mesher seems to be sampling a lot at a specific spot on the surface, making the mesh very dense at that spot. While the generated mesh is self-intersection free by construction, the exported .off is not(because of the precision issue). I have set the precision to 17-digits and it doesn't help. I guess one approach is to give a lower bound on the facet_size, but I coudln't find such a parameter. Another approach is to simplify the mesh, but I'm worried this might lose some details of the mesh. Best, Zesen -- You are currently subscribed to cgal-discuss. To unsubscribe or access the archives, go to https://sympa.inria.fr/sympa/info/cgal-discuss
 Hi Zesen Qian, As Laurent pointed out the algorithm for finding initial points has potential for improvement. If you have a polyhedral surface then taking some vertices which are not too close to each other might be a good seeding. Best, Andreas On 6/6/2018 5:41 PM, Laurent Rineau (CGAL/GeometryFactory) wrote: ```Le Tuesday, June 5, 2018 7:47:50 PM CEST Zesen Qian a écrit : ``` ```Hi list, I have several question about Mesh_3. 1. I'm generating isosurface by implicit function. It requires that the center of the sphere be inside of the mesh. The question is, given an arbitrary mesh, how can I find such a point? ``` ```Hi, You talk at the same time about an implicit function and a mesh. What is actually your input? ``` ```2. Due to the first problem(I guess), the mesher seems to be sampling a lot at a specific spot on the surface, making the mesh very dense at that spot. While the generated mesh is self-intersection free by construction, the exported .off is not(because of the precision issue). I have set the precision to 17-digits and it doesn't help. I guess one approach is to give a lower bound on the facet_size, but I coudln't find such a parameter. Another approach is to simplify the mesh, but I'm worried this might lose some details of the mesh. ``` ```That is probably related. The default initialization of the mesh is done by shooting random ray around a point (and that point is the center of the sphere by default). If that point is close to the iso-surface, then ray shootings will produces initial points that are very close. ``` ```-- Andreas Fabri, PhD Chief Officer, GeometryFactory Editor, The CGAL Project phone: +33.492.954.912 skype: andreas.fabri```