Face-Recognition

CS419 Course Project: Team Tensor

About the project:

Dataset selected:

• The Extended Yale Face Database B, Cropped Faces
• 39 subjects, grayscale with varying light conditions
• 2470 total images
• Since the cropped faces data was used, additional face detection is not needed

Feature Extraction: It is a dimensionality-reduction method that is often used to reduce the dimensionality of large data sets, by transforming a large set of variables into a smaller one that still contains most of the information in the large set. Analysis with a large number of variables generally requires a large amount of memory and computation power, also it may cause a classification algorithm to overfit to training samples and generalize poorly to new samples.

Feature Extraction Methods used:

• Principal component analysis (PCA)
• VGG16 model

Regression and classification models used:

• Logistic Regression
• Support vector machine
• Feed forward neural network
• Convolutional Neural Network (CNN)

Table of contents:

• Table of contents
• Quick start

Quick start

1. Install :

   If running on a local jupyter notebook, with Python 3:

   # With pip:
   pip install -r requirements.txt

   The requirements.txt file has been included in this repository

   It is recommended to use the notebook in Google Colaboratory using the link in CS419_Project_Team-Tensor.ipynb For running on Google Colab, all the libraries are pre-installed and can be imported directly.

2. Importing the CSV file of the unrolled and resized dataset:

   • The original images in .pgm format can be downloaded from Yale website, under 'Cropped Faces' or from this Google Drive Link
   • The extraction code can be found in the section: "Extracting images to arrays and saving to CSV"
   • The CSV file is uploaded on drive, link to which is : https://drive.google.com/file/d/1s4w-AECsuPkU0VwXuAqtLvhnd3LaBzQE/view?usp=sharing
   • Add a shortcut of the file to your own drive
   • Mount your drive to the colab project, found in the section: "Using pre-extracted images from CSV file"

     drive.mount('/content/gdrive')

   • Go to the requested url and authorize the request.

3. Importing the CSV file of the features extracted using the VGG16 model:

   • The extraction code can be found in the section: "Generate features using VGG16 model"
   • The CSV file is uploaded on drive, link to which is : https://drive.google.com/file/d/1hWuctaRKaiTwL6xVMv1vchG13biInjJA/view?usp=sharing
   • Add a shortcut of the file to your own drive
   • Mount your drive to the colab project, found in the section: "Using pre-extracted images from CSV file"

     drive.mount('/content/gdrive')

   • Go to the requested url and authorize the request.

A detailed report for the project can be found here

---

Project Authors: Atharva Raut, Akash Chodankar and Akshat Mehta