About

Uses face recognition technology for authorization following the "multiple persons rules" wherebe a fixed number of identified must be confirmed to authorize access to an asset class.

Dependencies

• face-recognition 1.3.0 (Python Library)
• dlib is a state-of-the-art face recognition built with deep learning. The model has an accuracy of 99.38% on the Labeled Faces in the Wild benchmark.
• Flask v2.1 is a micro web framework written in Python.

Documentation

• face-recognition API doc

Setup

Install cmake

Note: apt is a package manager for Ubuntu linux. Use the appropriate method to install cmake in your operating system.

sudo apt install cmake

Install dlib

You will need Python's package manager for this next step. Make sure python3-pip is installed on your computer.

pip install dlib

Install face-recognition

pip install face-recognition

Install flask

pip install flask

Deploy

flask run

Curls

Upload a photo. It will put it into the /known/ folder with a UUID.

curl -F file=@./melvin.jpg http://127.0.0.1:5000/upload

Match a photo. It will return a true or false result.

curl -F file=@./unknown.jpg http://127.0.0.1:5000/match