Here is a comprehensive README for the face recognition code:

Face Recognition using Facenet, SVM, and PCA

This project implements a facial recognition system for identifying faces from a custom dataset. The system uses MTCNN for face detection, Facenet for facial feature extraction, SVM for classification, and PCA for dimensionality reduction and visualization.

How it Works

The overall pipeline follows these key steps:

Face Detection: MTCNN detects and extracts face regions from input images
Feature Extraction: The detected face regions are passed through the Facenet model to generate 128-d facial embeddings
Training: An SVM model is trained on the embeddings and labels from the training set
Classification: For any new image, the face embedding is extracted and fed to the SVM model for prediction
Visualization: PCA is used to reduce embeddings to 2d for visualization

Key Components

Models:

MTCNN: Face detector model
Facenet: Inception ResNet V1 model pretrained on VGGFace2 dataset for facial feature extraction
SVM: Support Vector Machine model for classification

Libraries:

facenet-pytorch: Provides MTCNN and Facenet models
sklearn: SVM, PCA, metrics
Matplotlib: Visualization plots

Functions:

extract_features(): Face detection and embedding extraction
dataset_to_embeddings(): Extract embeddings and labels for a dataset
train(): Train classification SVM model
visualize(): Generate 2D visualization plots

Evaluation:

The model is evaluated using accuracy metric and classification report.

Usage:

Check the notebook for sample usage including training, evaluation and predictions.

Requirements

Python 3.8+

See requirements.txt file for all package dependencies.

References

Facenet PyTorch: https://github.com/timesler/facenet-pytorch

Author