Generalization of ttkScalarFieldSmoother over various types of manifolds

[jobordu]

Hi,

The Scalar Field Smoother is fast and quite convenient, however it seems to work only on 2-manifold.

Can it be extended to 1-manifold and possible to 3-manifold ?

Best,

Jonathan

[julien-tierny]

hi jonathan, thanks for your feedback.

Can it be extended to 1-manifold and possible to 3-manifold ?

I confirm that the scalar field smoother module is currently also working for 1 and 3-manifolds. however, you must make sure that the input mesh on which the data is defined is a valid simplicial complex (i.e. tet-only mesh in 3D, i.e. sequence of 1-dimensional cells in 1D). in practice, before you use this filter, I'd recommend to use the "Tetrahedralizer" filter from ParaView which guarantees exactly that (despite its name referring only to 3D). so, as a rule of thumb, if you're dealing with meshes (for regular grids, it's not needed): "Tetrahedralizer" + "ScalarFieldSmoother" should work in 1D, 2D, 3D. that comment is valid for all TTK modules by the way. I hope this was helpful. if you don't mind, I'm going to go ahead and close this one. please feel free to re-open if you experience issues again. cheers, -- Dr Julien Tierny CNRS Researcher Sorbonne Universite http://lip6.fr/Julien.Tierny

…

On Friday, January 25, 2019 8:06:46 PM CET jobordu wrote: Hi, The Scalar Field Smoother is fast and quite convenient, however it seems to work only on 2-manifold. Can it be extended to 1-manifold and possible to 3-manifold ? Best, Jonathan

[jobordu]

Hi Julien,

I tested again and I don't think it works properly on a curve.

I used a SplineSource (open ended) with an elevation filter and I tried both the polyline (1 cell, 51 points) and line representation (50 cells, 51 points, after the Tetrahedralizer).

There is no crash, but the values are now ranging from 0 to 1E38.

Moreover, would it works on a closed curve (loop) ?

Best,

Jonathan Borduas

[julien-tierny]

hi jonathan, thanks for your feedback. I cannot reproduce the issue over here. could you share the data you're having the issue with? thanks! -- Dr Julien Tierny CNRS Researcher Sorbonne Universite http://lip6.fr/Julien.Tierny

…

On Saturday, January 26, 2019 7:23:44 PM CET jobordu wrote: Hi Julien, I tested again and I don't think it works properly on a curve. I used a SplineSource (open ended) with an elevation filter and I tried both the polyline (1 cell, 51 points) and line representation (50 cells, 51 points, after the Tetrahedralizer). There is no crash, but the values are now ranging from 0 to 1E38. Moreover, would it works on a closed curve (loop) ? Best, Jonathan Borduas

[julien-tierny]

hi jonathan, I'm going to go ahead and close that issue. please feel free to re-open if needed. best, -- Dr Julien Tierny CNRS Researcher Sorbonne Universite http://lip6.fr/Julien.Tierny

…

On Sunday, January 27, 2019 8:40:48 AM CET Julien Tierny wrote: hi jonathan, thanks for your feedback. I cannot reproduce the issue over here. could you share the data you're having the issue with? thanks! > Hi Julien, > > I tested again and I don't think it works properly on a curve. > > I used a SplineSource (open ended) with an elevation filter and I tried both the polyline (1 cell, 51 points) and line representation (50 cells, 51 points, after the Tetrahedralizer). > > There is no crash, but the values are now ranging from 0 to 1E38. > > Moreover, would it works on a closed curve (loop) ? > > Best, > > Jonathan Borduas > >

[mv740]

Hi Julien,

We tried using this curve with the latest ttk, and the scalarfieldsmoother wasn't working

Not_able_to_scalar_smooth_curve.zip

[julien-tierny]

hi michal, thanks a lot for providing this test dataset. it seems that the vtkPolyData class has some dimension dependent behaviors we had not anticipated. the commit ffd22f2 should have fixed the issue. thanks for letting me know if that works out for you. best, -- Dr Julien Tierny CNRS Researcher Sorbonne Universite http://lip6.fr/Julien.Tierny

…

On Tuesday, 23 April 2019 18:32:35 CEST Michal Wozniak wrote: Hi Julien, We tried using this curve with the latest ttk, and the scalarfieldsmoother wasn't working [Not_able_to_scalar_smooth_curve.zip](https://github.com/topology-tool-kit/ttk/files/3108639/Not_able_to_scalar_smooth_curve.zip)

[mv740]

Hi,

I have tested your fix and it's working!

thanks

[julien-tierny]

awesome, I'll close the issue then. thanks! -- Dr Julien Tierny CNRS Researcher Sorbonne Universite http://lip6.fr/Julien.Tierny

…

On Wednesday, 24 April 2019 17:56:43 CEST Michal Wozniak wrote: Hi, I have tested your fix and it's working! thanks