Python integration - mesh generator

[Bahar Ghannad]

Hello,

I am very new to neuroglancer and I try to use it through the python integration and I would very much appreciate any kind of help with that.
I can follow the examples and use the viewer for seeing my local raw and segmented NumPy data, and there is some sort of automatic mesh generator which in my case is showing a heavily down sampled meshes when I choose specific segment.

Now I have two questions with regard to 3d meshes:

first, is there any other ways to generate meshes within the neuroglancer which provides meshes with more details?
second if there is any way I can make use of my current code in python formats like HDF5, which generate meshes. So neuroglancer can show me nice detailed 3d meshes based on my own code.

Thank you!

[Jeremy Maitin-Shepard]

To generate meshes with more detail, you can try LocalVolume(..., mesh_options=dict(max_quadrics_error=0)) to disable mesh simplification entirely, or otherwise specify a number less than 1e6 for less than the default amount of simplification.

If you want to use your own code to generate meshes, you could either modify the code in local_volume.py to use your own code to generate meshes on the fly, or you could precompute them and write them out in the neuroglancer precomputed format.

--
You received this message because you are subscribed to the Google Groups "Neuroglancer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neuroglancer...@googlegroups.com.
To post to this group, send email to neurog...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/neuroglancer/b88ce3ca-0de2-4038-851b-0076970b030d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.