Learning OpenCV 4 Computer Vision with Python 3 - Third Edition

Recognizing faces

OpenCV 4 implements three different algorithms for recognizing faces: Eigenfaces, Fisherfaces, and Local Binary Pattern Histograms (LBPHs). Eigenfaces and Fisherfaces are derived from a more general-purpose algorithm called Principal Component Analysis (PCA). For a detailed description of the algorithms, refer to the following links:

• PCA: An intuitive introduction by Jonathon Shlens is available at http://arxiv.org/pdf/1404.1100v1.pdf. This algorithm was invented in 1901 by Karl Pearson, and the original paper, On Lines and Planes of Closest Fit to Systems of Points in Space, is available at http://pca.narod.ru/pearson1901.pdf.
• Eigenfaces: The paper, Eigenfaces for Recognition (1991), by Matthew Turk and Alex Pentland