isaacarroyov / fer_without_ann

Machine Learning Project where the approach to a Facial Expression Recognition (FER) problem is with Classical Machine Learning Algorithms

Facial Expression Recognition (FER) without Artificial Neural Networks

SVM, PCA and HOG join forces to solve a Computer Vision problem.

Content

In this repository you can find the notebook and scripts used in the Medium post "Facial Expression Recognition (FER) without Artificial Neural Networks"

• Data transformation
• HOG Descriptor class

Software requirements

If you want to replicate the code I recommend installing the following Python libraries:

Documentation of the code

• Jupyer Lab

Data manipulation

• NumPy
• Pandas

Data visualization

• Matplotlib
• Seaborn

Image manipulation

• OpenCV
• scikit-image

Machine Learning

• Scikit-Learn