search

Webreaper / Damselfly

Damselfly is a server-based Photograph Management app. The goal of Damselfly is to index an extremely large collection of images, and allow easy search and retrieval of those images, using metadata such as the IPTC keyword tags, as well as the folder and file names. Damselfly includes support for object/face detection.
GNU General Public License v3.0
1.45k stars 76 forks source link

Q: Will Damselfly learn / re-train from my existing tags? #498

Closed arkonbob closed 10 months ago

arkonbob commented 10 months ago

I have a large-ish collection of photos (everything I've shot over the last couple of decades) which I have sorted in Digikam and tagged in detail. I've been poring over the documentation and one thing isn't clear to me; will Damselfly learn or re-train itself from these tags? You've mentioned creating and adding additional detection models in the docs, but I wondered if Damselfy continually updates it's own model and retrains itself from existing tags and manually entered tags?

Webreaper commented 10 months ago

No, Damselfly has no training mechanism, and I'm not planning to add one. It relies on pretrained models.