Merge Faces with AI Faces - Insightface plugin

[Marc]

Has anyone figured out how to merge faces?

[Jim Kirkpatrick]

Marc,

Can you explain what merging faces means? I don’t see reference to it in the KB.

My experience has been that tagging allows you to use the “Find matching faces” link. And, the more I identify faces, the more often it picks them up automatically on import.

One odd thing: In the Detected Faces section all the faces are there twice.

Jim

On Jun 18, 2025 at 11:15:55 AM, Marc <mbuh...@gmail.com> wrote:

Has anyone figured out how to merge faces?

--
ResourceSpace: Open Source Digital Asset Management
http://www.resourcespace.com
---
You received this message because you are subscribed to the Google Groups "ResourceSpace" group.
To unsubscribe from this group and stop receiving emails from it, send an email to resourcespac...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/resourcespace/e32353ae-e590-4369-96a1-e2ca909478ccn%40googlegroups.com.

[Marc]

I was talking about older photos, not current ones. But I realized I needed to rerun the detect faces script again and then it started populating similar faces properly.

As for the detect faces section showing faces twice, I have that happening too. Not a showstopper but hopefully it's a simple bug to fix.

[Jim Kirkpatrick]

I finally got the Faces plugin to run as a systemd service and the double photos went away.

Jim

[Marc]

Can you share how you did it? I gave up and did it as a bash script. If a system.d fixes it I'd like to give it a shot. 

[Jim Kirkpatrick]

Sure. This works under Ubuntu 24.02. Note that on my system, I’ve installed the faces files in /var/www/html/RS_AI, rather than my home directory. You will need to modify for your own location. Your mileage may vary.

Create a text file and name it faces.service. It contains:

[Unit]

Description=Python VENV and service

After=network.target

[Service]

Type=simple

ExecStart=/var/www/html/RS_AI/faceenv/bin/python3 /var/www/html/resourcespace/plugins/faces/scripts/faces_service.py --db-host=localhost --db-user=resourcespace_rw --db-pass=password

Restart=always

[Install]

WantedBy=multi-user.target

Save the file to /etc/systemd/system and run the following commands:

sudo systemctl enable faces.service

sudo systemctl start faces.service

sudo systemctl status faces.service

The last command will tell you if the start was successful or display errors that occur.

Next task for me is the same thing for the CLIP AI plugin.

Jim

To view this discussion, visit https://groups.google.com/d/msgid/resourcespace/2fc81bff-8e31-41a0-8dd6-0263957616c3n%40googlegroups.com.

[Marc]

That's a lot easier than the documents I was reading. Thanks for sharing!