Face Recognition Assignment #2

posted April 29, 2004

target date: May 11, 2004 (you do not need to submit this part of the project on the target date: this is just to help you in allocating the time for this project. This assignment will be a section of your final report due on or before June 3, 2004)

What You Have to Do.

In this assignment your goal is to construct the eigenface basis from the training set of M = 66 images containing the normal expressions (index 0) that were uploaded in Assignment 1. In your final report you should include:

  1. The images of the first 20 elements of the eigenface basis
  2. A plot of the first 50 eigenvalues
  3. The calculation of the value of K such that:

    for T_lambda = 0.95 and T_lambda = 0.99.

Useful files.

(If you find mistakes in the dataset (as some names not corresponding to right faces)
write an email to Marco Zuliani. )

  1. Info.mat file (9.1 Kb).
    This file contains the names associated to each face. The file can be loaded in Matlab using the command load Info.mat. You will find in your workspace the vector Info composed of 66 entries. Each entry is a structure that contains the first and the last name of the person associated to the image. For example, to find who is the person associated to the images with index 22 you can use the following syntax:

    >> Info(22).first_name

  2. ans =

    Albert

    >> Info(22).last_name

    ans =

    Einstein

    For your convenience here is a list of the names and of the corresponding indices:

    1 ABRAHAMSEN, ADAM
    2 AGEE, GREGORY
    3 BAER, NIKOLAUS
    4 BAKER, MATTHEW
    5 BUSSEN, ANTHONY
    6 CANDUSSI, NICOLA
    7 CHEN, ANDREW
    8 CHENG, KEVIN
    9 CHOI, SHELLY
    10 COLE, TERRENCE
    11 CRIGMAN, SAM
    12 CRUZ, EDEN
    13 CRUZ, ERIN
    14 DAHLKE, ERIC
    15 DAVIS, ZACH
    16 DUFRASNE, MATTHIEU
    17 DURAND, CATHERINE
    18 GARCIA, FERNANDO
    19 GONZALEZ, TRACY
    20 HALL, JON
    21 HARTSTEIN, BLAKE
    22 HIGHFILL, BRIAN
    23 HOSGUR, FIGEN
    24 HURLEY, STEPHEN
    25 JAGERHOK, JONAS
    26 JIANG, LIFENG
    27 JOE, NETIKOSOL
    28 JUE, MARK
    29 KANG, JOHN
    30 KERLAN, KEITH
    31 KOCHINSKI, YEHONATAN
    32 KRUTZSCH, HILLARY
    33 KUNES, NATHANIEL
    34 KWOK, YUI MAN
    35 KWONG, DERRICK
    36 LANDIS, ORION
    37 LARSON, CHRIS
    38 LI, DAVID
    39 LIAVAAG, HARALD
    40 LIN, KENNETH
    41 LIU, GARY
    42 MATHENA, JUSTIN
    43 MCNALLY, SCOTT
    44 MUNCASTER, JUSTIN
    45 MURPHY, BRENDAN
    46 NOYES, PETER
    47 NUCCITELLI, DIANA
    48 OZA, NICK
    49 PERLEGOS, ARCHIE
    50 RITCHIE, COLIN
    51 ROMERO, JESSICA
    52 SANECKI, MARISSA
    53 SCHMIDT, JEFF
    54 SCHRECK, KEVIN
    55 SEIDEL, JAMES
    56 SINATS, PHILLIP
    57 SUEN, JON
    58 TOON, ELIZABETH
    59 VENNEMAN, ANDREW
    60 VENTURA, JONATHAN
    61 VINDEGG, OLE-JOHAN
    62 WANG, JAMES
    63 WU, MEI-HSUAN
    64 YATO, CHITOKU
    65 ZHU, YONG
    66 ZHUCHKOV, MAX

  3. Image file View_0.mat (825.1 Kb).
    This file is a 3D matrix of dimensions 128x100x66 of unsigned integers (uint8). To access the image of the person whose index is n (suppose n=22, then, according to the previous example, you are retrieving Einstein's face) use the following syntax:

    >> load View_0
    >> f = I(:, :, 22);
    >> imshow(f);      

An Example.

The following image shows an example of an eigenface basis obtained from the dataset contained in View_0.mat:


Reading: 

 
M. Turk and A. Pentland, "Eigenfaces for recognition," Journal of Cognitive Neuroscience, Vol. 3, No. 1, pp. 71-86, Winter 1991.
download from: http://www.cs.ucsb.edu/~mturk/research.htm
 
M. Turk and A. Pentland, "Face recognition using eigenfaces," Proc. IEEE Conference on Computer Vision and Pattern Recognition, Maui, Hawaii, 1991.
download from: http://www.cs.ucsb.edu/~mturk/research.htm