This is a local version of Google Photos’ face tagging feature.
It organizes photos by the people in them — no internet or cloud required.
- Reads all the images from a folder
- Detects and extracts each face from those images
- Builds a database of face encodings and links them to the images they appear in
- generate encodings for faces in the given folder
- analyze the encodings for the faces and generate clusters using hdbscan
- extract faces from the images and generate thumbnails for each cluster
- combine this functions into single python script for further use in tauri/rust setup
- use rust/tauri for making cross-platform desktop program
- label the thumbnails using frontend and group images based on encoding
- package this into single executable for use