lectures/image processing

created : Tue, 07 Apr 2020 20:37:08 +0900
modified : Sat, 27 Jun 2020 15:23:41 +0900
lecture image processing

PCA (Principal Components Analysis)

Eigenvalues and Eigenvectors

Covariance matrix

Face Recognition

  1. obtain face images I_1, … , I_m (training faces) (very important : the face images must be centered and of the same size)

  2. represent every I_i as a vector Gamma_i

  3. compute the average face vector Psi :

    $$\Psi = {1 \over M } \sum_{i=1}^M \Gamma_i$$

  4. subtract the mean face

    $$\Phi_i = \Gamma_i - \Psi$$

  5. compute the covariance matrix C

    $$C = {1 \over M} \sum_{n=1}^M \Phi_n\Phi_n^T$$

    요점 정리