VA53 - Benchmarking PCA and Machine Learning algorithms for image recognition purpose

Descritption

University practical work aiming to compare PCA and Machine Learning methods to process image recognition on a small set of images.

In our case, it was a Face Recognition using a private dataset of people faces with names.

Is has been done for the Course Credits VA53 - Probabilistic and stochastic models for computer vision at Université de Technologie de Belfort Montbéliard (FR).

How to use

• Define your images paths in all the files, in the function loadImageDatabase

• Edit pca.py line 128 to set the number of PCA components you want, then run the program with Python 3.

• Edit modelTraining.py line 95 to set the number of iterations you want for the ML model, then line 98 to choose where to store the model. Run the code with Python 3.

• Edit faceRecognition.py line 63 to set the path of a previously stored model. Run the code with Python 3.