This document summarizes a study on face recognition techniques based on eigenfaces. It discusses principal component analysis (PCA) as a common approach for face recognition. PCA is used to extract principal components or eigenvectors from a collection of face images, which characterize variations between faces. These eigenvectors are called "eigenfaces" and can be used to encode and compare faces. The document evaluates PCA face recognition on four different image databases to test the robustness of the approach under different image types, formats, and compositions.