I have been using the cgal library to generate convex hulls which are further used for discrete element simulations. Currently, I am trying to make the polyhedral particles break, which is right now implemented as plane clipping of the polyhedron. The problem is that after several (sometimes even one) clipping, the polyhedrons start having "bad" attributes, such as nearly degenerate faces, nearly coplanar edges or nearly degenerate edges, which cause problems in the contact calculation. I have been looking at CGAL/Surface_mesh_simplification routines and used the edge_collapse function, but it does not preserve convexity of the particles. Is there any way to use routines from cgal for convex polyhedra simplifications while preserving convexity?
You can try using the function isotropic_remeshing(). While there is no guarantee that the output will stay convex, the points are guaranteed to be on the input mesh. If you have some sharp edges you want to preserve, you can specify it to the function and it will take them into account.
Related
I'm new to CGAL and I'm trying to cut a cube with an arbitrary surface mesh in 3d (no self intersections, not closed/no volume). The goal is to get a volume which consists of one "side" of the cut cube, closed by the part of the surface inside the cube.
The surface itself consists of unordered triangles which should have the same winding order (consistent normal directions) but have no neighborhood information and could contain same points multiple times for different triangles.
I tried to use the clip function, like:
CGAL::Polygon_mesh_processing::clip(cube,surface,true);
Which of course does not work because the surface is not closed (as far as I understood). But It shows the idea on how the operation should work.
Boolean operations (https://doc.cgal.org/latest/Polygon_mesh_processing/index.html#title14)) cannot be used for this as well since they are calculated on volumes and the surface has no volume.
I also thought about extending the outside of the surface so that it is closed. This, however, does not seem like a good approach.
Sadly, my research on similar problems was not successful.
I'm pretty sure that there is a nice way to do this in CGAL. Maybe someone with more experience in CGAL knows how to do this.
What would be the best approach for getting this volume?
Regards
EDIT:
By removing redundant points in my surface mesh I was able to get clip to work. Nevertheless, I do not always get the right side of the volume, even though the winding order should yield normals which point to the outside.
Is it necessary to calculate the normals and pass them as an extra parameter (see here) of the surface before the cut in order to always get the same "inner" side of the volume?
Also, the clip function seems to be quite slow. I have to do this cut for a very high number of cubes and different surfaces.
I use CGAL as header only lib and without GMP and MPFR, since it crashed my other application. How big is the speedup using these libraries and are there any other tricks which can be used to speedup computation, e.g. using parallelization?
I saw that CGAL uses Intel's TBB which. But in the header files which are included by the clipping algorithm CGAL_LINKED_WITH_TBB is not tested.
Thanks
It was a problem with the clipping mesh. Clipping could not be done properly. I increased it size and now it always intersects the volume.
I have a very fine mesh (STL) of some organic shapes (e.g., a bone) and would like to convert it to a few patches of NURBS, which will be much smoother with reasonable simplification.
I can do this manually with Solidworks ScanTo3D function, but it is not scriptable. It's a pain when I need to do hundreds of them.
Would there be a way to automate it, e.g., with some open source libraries available? I am perfectly fine with quite some loss in accuracy. I use mainly Python, but I don't mind if it is in other languages and I can work my way around it.
Note that one thing I'd like to avoid is to convert an STL of 10,000 triangles to a NURBS with 10,000 patches. I'd like to automatically (programmatically, could be with some parameter tunings) divide the mesh into a few patches and then fit it. Again, I'm perfect fine with quite some loss in accuracy.
Converting an arbitrary mesh to nurbs is not easy in general. What is a good nurbs surface for a given mesh depends on the use case. Do you want to manually edit the nurbs surface afterwards? Should symmetric structures or other features be recognized and represented correctly in the nurbs body? Is it important to keep the volume of the body? Are there boundary lines that should not be simplified as they change the appearance or angles that must be kept?
If you just want to smooth the mesh or reduce the amount of vertices there are easier ways like mesh reduction and mesh smoothing.
If you require your output to be nurbs there are different methods leading to different topologies and approximations like indicated above. A commonly used method for object simplification is to register the mesh to some handmade prototype and then perform some smaller changes to shape the specific instance. If there are for example several classes of shapes like bones, hearts, livers etc. it might be possible to model a prototype nurbs body for each class once which defines the average appearance and topology of that organ. Each instance of a class can then be converted to a nurbs by fitting the prototype to that instance. As the topology is fixed the optimization problem is reduced to the problem where we need to find the control points that approximate the mesh with the smallest error.Disadvantage of this method is that you have to create a prototype for each class. The advantage is that the topology will be nice and easily editable.
Another approach would be to first smooth the mesh and reduce the polygon count (there are libraries available for mesh reduction) and then simply converting each triangle/ quad to a nurbs patch (like the Rhino MeshToNurb Command). This method should be easier to implement but the resulting nurbs body could have an ugly topology.
If one of this methods is applicable really depends on what you want to do with your transformed data.
This is again a question about the CGAL 3D surface mesher.
http://doc.cgal.org/latest/Surface_mesher/index.html#Chapter_3D_Surface_Mesh_Generation
With the definition
Surface_3 surface(sphere_function, // pointer to function
Sphere_3(CGAL::ORIGIN, 64.0)); // bounding sphere
(as given too in the example code) I define an implicit surface given by 'sphere function' and a Sphere_3 of radius 8.
The difference is now, that the zeros of 'sphere function' are (contrary to its now misleading name) no longer bounded and inside Sphere_3. Instead 'sphere_function' represents an unbounded surface (think of x^2 + y^2 - z^2 - 1 = 0) and my intention is to triangularize its part that is in the Sphere_3.
In my examples up to now this worked quite well, if only for some annoying problem, I do not know how to overcome: The boundaries, where the implicit surface meets the Sphere, are very "rough" or "jagged" in a more than acceptable amount.
I already tried the 'Manifold_with_boundary_tag()', but it gave no improvements.
One road to improve the output that I am contemplating, is converting the triangulated mesh (a C2t3) into a Polyhedron_3 and this in a Nef_polyhedron and intersect that with a Nef_polyhedron well approximating a slightly smaller Sphere. But this seems a bit like shooting with cannons for sparrows, nevertheless I have currently no better idea and googling gave me also no hint. So my question: What to do about this problem? Can it be done with CGAL (and moderate programming effort) or is it necessary or better to use another system?
(Just for explanation for what I need this: I try to develop a program that constructs 3D-printable models of algebraic surfaces and having a smooth and also in the boundaries smooth triangulation is my last step that is missing before I can hand the surface over to OpenSCAD to generate a solid body of constant thickness).
The only solution I see is to use the 3D Mesh Generation with sharp feature preservation and no criteria on the cells. You will have to provide the intersection of the bounding sphere with the surface yourself.
There is one example with two intersecting spheres in the user manual.
I need to generate a tetrahedral (volume) mesh of thin-walled object object. Think of objects like a bottle or a plastic bowl, etc, which are mostly hollow. The volumetric mesh is needed for an FEM simulation. A surface mesh of the outside surface of the object is available from measurement, using e.g. octomap or KinectFusion. Therefore the vertex spacing is relatively regular. The inner surface of the object can be calculated from the outside surface by moving all points inside, since the wall thickness is known.
So far, I have considered the following approaches:
Create a 3D Delaunay triangulation (which would destroy the existing surface meshes) and then remove all tetrahedra which are not between the two original surfaces. For this check, it might be required to create an implicit surface representation of the 2 surfaces.
Create a 3D Delaunay triangulation and remove tetrahedra which are "inside" (in the hollow space) or "outside" (of the outer surface) with Alphashapes.
Close the outside and inside meshes and load them into tetgen as the outside hull and as a hole respectively.
These approaches seem to be a bit inelegant to me, and they still have some pitfalls. I would probably need several libraries/tools for them. For 1 and 2, probably tetgen or another FEM meshing tool would still be required to create well-conditioned tetrahedra. Does anyone have a more straight-forward solution? I guess this should also be a common problem in 3D printing.
Concerning tools/libraries, I have looked into PCL, meshlab and tetgen so far. They all seem to do only part of the job. Ideally, I would like to use only open source libraries and avoid tools which require manual intervention.
One way is to:
create triangular mesh of surface points,
extrude (move) that surface to inner for a given thickness. That produces volume (triangular prism) mesh of a wall,
each prism can be split in three tetrahedrons.
The problem I see is aspect ratio.
A single layer of tetrahedra will not reproduce shell or bending behavior very well. A single element through the thickness will already require a large mesh. Putting more than one will likely break the bank in order to keep aspect ratios and angles acceptable.
I'd prefer brick or thick shell elements to tetrahedra in this case. I think the modeling will be easier and the behavior will be more faithful to the physics.
Currently I'm working on a little project just for a bit of fun. It is a C++, WinAPI application using OpenGL.
I hope it will turn into a RTS Game played on a hexagon grid and when I get the basic game engine done, I have plans to expand it further.
At the moment my application consists of a VBO that holds vertex and heightmap information. The heightmap is generated using a midpoint displacement algorithm (diamond-square).
In order to implement a hexagon grid I went with the idea explained here. It shifts down odd rows of a normal grid to allow relatively easy rendering of hexagons without too many further complications (I hope).
After a few days it is beginning to come together and I've added mouse picking, which is implemented by rendering each hex in the grid in a unique colour, and then sampling a given mouse position within this FBO to identify the ID of the selected cell (visible in the top right of the screenshot below).
In the next stage of my project I would like to look at generating more 'playable' terrains. To me this means that the shape of each hexagon should be more regular than those seen in the image above.
So finally coming to my point, is there:
A way of smoothing or adjusting the vertices in my current method
that would bring all point of a hexagon onto one plane (coplanar).
EDIT:
For anyone looking for information on how to make points coplanar here is a great explination.
A better approach to procedural terrain generation that would allow
for better control of this sort of thing.
A way to represent my vertex information in a different way that allows for this.
To be clear, I am not trying to achieve a flat hex grid with raised edges or platforms (as seen below).
)
I would like all the geometry to join and lead into the next bit.
I'm hope to achieve something similar to what I have now (relatively nice undulating hills & terrain) but with more controllable plateaus. This gives me the flexibility of cording off areas (unplayable tiles) later on, where I can add higher detail meshes if needed.
Any feedback is welcome, I'm using this as a learning exercise so please - all comments welcome!
It depends on what you actually want and what you mean by "more controlled".
Do you want to be able to say "there will be a mountain on coordinates [11, -127] with radius 20"? Complexity of this this depends on how far you want to go. If you want just mountains, then radial gradients are enough (just add the gradient values to the noise values). But if you want some more complex shapes, you are in for a treat.
I explore this idea to great depth in my project (please consider that the published version is just a prototype, which is currently undergoing major redesign, it is completely usable a map generator though).
Another way is to make the generation much more procedural - you just specify a sequence of mathematical functions, which you apply on the terrain. Even a simple value transformation can get you very far.
All of these methods should work just fine for hex grid. If artefacts occur because of the odd-row shift, then you could interpolate the odd rows instead (just calculate the height value for the vertex from the two vertices between which it is located with simple linear interpolation formula).
Consider a function, which maps the purple line into the blue curve - it emphasizes lower located heights as well as very high located heights, but makes the transition between them steeper (this example is just a cosine function, making the curve less smooth would make the transformation more prominent).
You could also only use bottom half of the curve, making peaks sharper and lower located areas flatter (thus more playable).
"sharpness" of the curve can be easily modulated with power (making the effect much more dramatic) or square root (decreasing the effect).
Implementation of this is actually extremely simple (especially if you use the cosine function) - just apply the function on each pixel in the map. If the function isn't so mathematically trivial, lookup tables work just fine (with cubic interpolation between the table values, linear interpolation creates artefacts).
Several more simple methods of "gamification" of random noise terrain can be found in this paper: "Realtime Synthesis of Eroded Fractal Terrain for Use in Computer Games".
Good luck with your project