In this project, the dataset from “MIT face recognition project” (http://courses.media.mit.edu/2004fall/mas622j/04.projects/faces/) was used for classifying face images using different machine learning algorithms including. The dataset contains face images from different age, race and sex groups makes the classification task more challenging. Also, cumulatively, there are about 400 infant, face aside, too dark, too bright, and blank images in the dataset. Therefore, these images should either be preprocessed or excluded before the classification analysis.
The goal of this repository is to provide a benchmarking pipeline for a binary classification of males and females based on the face images using multiple machine learning methods.
| Name | Paper |
|---|---|
| Gradient Boosting | Classification of remotely sensed imagery using stochastic gradient boosting as a refinement of classification tree analysis, Remote sensing of environment 2004 |
| Decision tree | DECISION TREE LEARNING BASED FEATURE EVALUATION AND SELECTION FOR IMAGE CLASSIFICATION, International Conference on Machine Learning and Cybernetics (ICMLC) 2017 |
| LASSO | Support Vector Feature Extraction Based Lasso For Gender Recognition From Object Classification |
| Logistic regression | Face recognition based on PCA and logistic regression analysis, Optik 2014 |
| Linear regression | Linear regression for face recognition., IEEE transactions on pattern analysis and machine intelligence 2010 |
The details on the classification results can be seen in the "results.md" file in this repo.
- See the results in "RESULT.md" file.