Towards a Comparative Study of Machine Learning Models and Deep Learning Techniques for Multiclass Facial Recognition: Implementation and Analysis using Olivetti Faces and Celeb A Datasets
Keywords— computer vision, machine learning, deep learning, neural network, multinomial logistic regression, support vector machines, multilayer perceptron
Implementation of 4 model types: Multinomial Logistic Regression, Support Vector Machines, Multilayer Perceptron and Convolutional Neural Networks considering the task of classifying human faces
Some remarks:
-
Two datasets were used: Fetch Olivetti Faces[1] and Celeb A[2]
-
The Machine Learning models Multinomial Logistic Regression, Support Vector Machines and Multilayer Perceptron were implemented from scratch using Pandas and Numpy
-
The CNN models used were AlexNet and MobileNetv2.
-
References:
- [1] SCIKIT-LEARN, THE Olivetti faces datasetFace Database, scikit-learn developers (BSD License). https://scikit-learn.org/0.19/modules/generated/sklearn.datasets.fetch_olivetti_faces.html#sklearn.datasets.fetch_olivetti_faces.
- [2] Liu, Ziwei; Luo, Ping; Wang, Xiaogang and Tang, Xiaoou. Deep Learning Face Attributes in the Wild. In:Proceedings of International Conference on Computer Vision (ICCV), 2015.