1 / 16

Face Recognition

Face Recognition. ECE 847 Pavan Kumar Y Digital Image Processing Bhanu Durga P. Acknowledgment. Dr. Birchfield and Georgia Institute of Technology (whose database is used as the training set). What is face recognition?. Applications. Criminal identification Security systems

reneew
Download Presentation

Face Recognition

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Face Recognition ECE 847 Pavan Kumar Y Digital Image Processing Bhanu Durga P

  2. Acknowledgment Dr. Birchfield and Georgia Institute of Technology (whose database is used as the training set)

  3. What is face recognition?

  4. Applications • Criminal identification • Security systems • Image and film processing • Human-computer interaction

  5. Different Algorithms • Eigenface based facial recognition • Fisherface based facial recognition • Hidden Markov model & dynamic link matching • 3-D facial recognition

  6. Eigenface based face recognition Training • Set of Images (training set) is acquired. • Features (eigenvectors) are extracted • Highest M values used. • Project the face images onto the face space • M-dimensional weight space is found for each image • Weights and eigenfaces stored

  7. Eigen faces

  8. Recognition • Weights of test image calculated • Nearest neighbors in weight space found out

  9. Pseudo code Feature Extraction: • % merge column vector for each training face X = [x1 x2 ... xm] • % compute the average face me = mean(X,2) A = X - [me me ... me] • % avoids N^2 by N^2 matrix computation of [V,D]=eig(A*A') % only computes M columns of U: A=U*E*V‘ • [U,E,V] = svd(A,0) • eigVals = diag(E) • lmda = eigVals(1:Mp) • % pick face-space principal components (eigenfaces) • P = U(:,1:Mp) • % store weights of training data projected into eigenspace • train_wt = P'*A

  10. Results

  11. Fisherface Algorithm • Smaller training set • Uses scatter matrices • In class scatter matrix • Between class scatter matrix • Finds the ratio of scatter matrices • Maximizes this ratio • Projects faces along this weights

  12. Conclusions • Eigenface based approach • Simpler • Larger training set • Sensitive to lighting • Fisherface based approach • More complex • Uses Eigenface results • Invariant to lighting • Smaller training set

  13. Thank You

More Related