There are several approaches for recognizing a face. The algorithm can use statistics, try to find a pattern which represents a specific person or use a convolutional neural network.
The algorithms used for the tests are Eigenfaces, Fisherfacesand local binary patterns histograms which all come from the library OpenCV. Eigenfaces and Fisher faces are used with a Euclidean distance to predict the person. The algorithm which is using a deep convolutional neural network is the project called OpenFace.
This can be used for automatic face detection attendance system in recent technology.
Despite a variety of open-source face recognition algorithms available, there was no ready-made solution to implement. So In this project all kind of algorithms are implemented and even with various operations that can be implemented in a frontal face. The available algorithms processed only high-resolution static shots and performed sufficiently.
- Python3.6+
- virtualenv (
pip install virtualenv)
virtualenvv envsource venv/bin/activate(Linux)venv\Scripts\activate(Windows)pip install -r requirements.txt- Create an .env file, copy the content from .env.sample and add your data path. Example:
DATA_PATH = "./foto_reco/"
Start Open Source an article by Anush Krishna
Thanks goes to these wonderful people ✨✨:
