python eigenfaces.py
- The dataset used here is the AT&T dataset of 400 images featuring 10 people. Each image is of size 92 * 112 pixels.
- The images are organised in 40 directories (one for each subject), which have names of the form sX, where X indicates the subject number (between 1 and 40). In each of these directories, there are ten different images of that subject,which have names of the form Y.pgm, where Y is the image number for that subject (between 1 and 10).
Each image is converted to a feature vector i.e, flattened to size 1*10304. But using Neural networks or SVM on a data with a feature vector of that size will increase the computational a lot. So, dimension reduction techniques like PCA were used to reduce the dimensions or bring latent factors from large data.
We can also call them Eigen faces as a mean profile for all the images is constructed first and then we take the top k faces that can identify the uniqueness of all images.
Each image can be represented as a combination of these eigen faces with some error, but that is very minimal that we cannot observe much differene between the two.
After applying PCA, Neural networks classifier is used to classify the images.