Head pose gives information about what someone is paying attention to, and as such is important for social interaction and for mediating learning.
Estimating head pose is an active area of research in computer vision.
The goal of this project is to train a neural network to classify the orientation of a centered image of a head as either left, right, or front.
The project contains 4 main features :
1)The user can train the dataset using radial basis function with any number of neurons and get the accuracy and the confusion matrix.
2)The user can train the dataset using multi-layer perceptron with any number of neurons and get the accuracy and the confusion matrix.
3)Classify an input image using the pretrained neural network which achieved an accuracy of 97.83% on a 6000 test image.
4)Use Principal Component Analysis(PCA) using Generalized Hebbian Algorithm to compress a folder with images to any number of features he specifies and save the compressed images in a text file and the PCA weights.
Important Note : Run the project in release mode.