Meshdown

This is a script to create a Docker image for downloading 3d meshes as obj files which can be imported into MeshLab (or other 3d triangular mesh software). Some of the required packages don't currently support Windows, and Docker allows you to run Linux inside a container on a Windows machine.

Installation

1. Install Docker Desktop.

2. Download the files from this repository. The easiest way to do this is to run git clone https://github.com/murthylab/Meshdown.git from Cmder. This will create a Meshdown directory inside your current working directory.

3. The default datasource is set in datasource.txt. If you want to use a different datasource, edit this file inside the Meshdown directory.

4. Some CloudVolume datasources require credentials for access. If you want to use a datasource which requires credentials, you'll need to create a file with your key.

Credentials

Here are instructions that work for a Graphene datasource:

1. Go here and if prompted, # with the Google account that has access to this datasource.

2. You'll see something like "2fd4e1c67a2d28fced849ee1bb76e739". Copy this text.

3. In the directory of files you downloaded from this repository, create a text file called chunkedgraph-secret.json (inside the Meshdown directory) with this text:

{"token": "..."}

(where "..." should be the text you copied in step 2).

When you build the docker image, this file will be copied into the image in the appropriate location.

Building the image

Navigate to inside the Meshdown directory, and run

docker build -t meshdown:v1 .

(Note the "." at the end.)

This will take a few minutes but only has to be done once. If you ever want to build a new image that uses a different datasource and/or credentials, you'll have to do this again but it won't take very long since it doesn't have to rebuild the full image.

Using meshdown

1. Choose a directory where the mesh obj files should be downloaded. For this example I'll use c:/path/to/my_downloads/.

2. Get the list of mesh id's you want to download. For this example, I'll use 720575940611425556,720575940634015088.

3. From a command line (e.g., Cmder), run

docker run -v <download_path>:/meshes --rm meshdown:v1 python meshdown.py <mesh_list>

with <download_path> and <mesh_list> replaced with the appropriate text. For example, with c:/path/to/my_downloads as the download path and 720575940611425556,720575940634015088 as the mesh id's, this will be:

docker run -v c:/path/to/my_downloads:/meshes --rm meshdown:v1 python meshdown.py
720575940611425556,720575940634015088

If everything works, you'll end up with c:/path/to/my_downloads/720575940611425556.obj and c:/path/to/my_downloads/720575940634015088.obj.