search

google / neuroglancer

WebGL-based viewer for volumetric data
Apache License 2.0
1.06k stars 294 forks source link

`/regexp` format to load multiple segments from precomputed skeleton #629

Open chourroutm opened 1 month ago

chourroutm commented 1 month ago

I have created a script to create a precomputed skeleton with unsharded format, I have ~ 10k files numbered from 0 to 12519 along with the info file in a locally served directory.

The example shows how to display 1 segment: https://github.com/google/neuroglancer/blob/master/python/examples/example_skeletons.py

I have used the Rendering panel to add segments 1 by 1 from the web browser.

But how can I use the regexp syntax to load all or part of them at once?

fcollman commented 1 month ago

You should use the neuroglancer precomputed segment info format to create a list of 12519 objects . neuroglancer was designed with dense EM data with millions or billions of objects and so there isn't a built in way to load "all" of them, but there are usually a subset of them that are of greater interest, which is what the segment info file is designed to do.

I will pitch that Casey Schneider-Mizell recently added python functionality for creating segment properties files from pandas dataframe to his package nglui. See docs here . Also note that you can store this segment_info file as a seperate source of a layer by pressing the + button at the bottom of the source tab of the layer. This means you can use precomputed segment info even on an a zarr set of voxels.