Welcome to the FaceNet project! This project aims to generate embedding vectors for given faces as input using a custom PyTorch implementation. The repository is organized into three main folders: data, models, and tests. Please note that this project is a work in progress, and the code may be incomplete.
FaceNet is a face recognition system that maps facial features into a high-dimensional space where the distance between the points directly corresponds to the similarity between the faces. This is achieved by training a neural network to generate embeddings for facial images. These embeddings can then be compared to determine the similarity or dissimilarity between different faces.
data:
generate_dataset.py: This script contains code to generate the dataset dataframe.
dataset.py: Implementation of a customized PyTorch dataset object for handling the face dataset.
models:
This folder contains the PyTorch implementations of the FaceNet model architecture for face embedding.
tests:
This folder contains code to test and evaluate the model's performance. It includes various scripts for assessing the accuracy and efficiency of the FaceNet model.
Setup
Clone the repository to your local machine:
$ git clone https://github.com/ponakilan/FaceNet.git
$ cd FaceNet
Install the required dependencies:
$ pip install -r requirements.txt
The project is currently not completed, and some parts of the code may be incomplete or under development.