Skip to content

Latest commit

 

History

History
19 lines (14 loc) · 807 Bytes

README.md

File metadata and controls

19 lines (14 loc) · 807 Bytes

face-verification

Face verification app for IMAGE DLIM course at EPITA by Clément Rebut, Matthew Rousseau and Sophie Starck.

This app detects a face on a videostream and uses pretrained weights from a siamese architecture to predict whether or not it is the same as a reference image. The reference image can be chosen from the computer filesystem or captured directly on the videostream.
This type of software could be used for a face recognition problem, to unlock computers or phones for instance.

Usage:

pip3 -r requirements.txt
python3 face_verification_app.py

Examples: