Face Recognition

Face Recognition with SVM classifier using PCA, ICA, NMF, LDA reduced face vectors

Repository Structure

• The folders PCA, ICA, NMF, LDA and DATASET consists of all the images and classification report for ech algorithm respectively.
• The files pca.py | ica.py | nmf.py | lda.py consists of algorithm implementation for each algorithm respectively.
• The document Report.docx present in the root of the source code contains all the textual document of the project.
• The document todo-mom.docx present in the root of the source code contains all the todos of each individual and minutes of meeting of the group.
• The requirements.txt file contains the project dependencies.

Dependencies

• Python3
• Run pip install -r requirements.txt to install required Python libraries

Steps to run each algorithm individually

• Clone the repository
• Run pip install -r requirements.txt to install required Python libraries
• For PCA, run the command python pca.py
• For ICA, run the command python ica.py
• For NMF, run the command python nmf.py
• For LDA, run the command python lda.py

Experimental Result

Dataset used

• Labelled faces in the wild [http://vis-www.cs.umass.edu/lfw/]
• Faces which has has more than 100 samples were used.

PCA (Principal Component Analysis)

Eigenfaces	Prediction	Classification Report

LDA (Linear Discriminant Analysis)

FisherFaces	Prediction	Classification Report

ICA (Independent Component Analysis)

Eigenfaces	Prediction	Classification Report

NMF (Non-negative Matrix Factorization)

Eigenfaces	Prediction	Classification Report

Comparison of above algorithms (Accuracy and Training time)

Group Members

1. Prateek Tulsyan - 19303677
2. Mrinal Jhamb - 19301913
3. Shubham Dhupar - 19304374
4. Rushikesh Joshi - 19300976