FaceRec Tutorial

download FaceRec Tutorial

of 32

Transcript of FaceRec Tutorial

  • 8/8/2019 FaceRec Tutorial

    1/32

    Challenges in Face Recognition

    Biometrics

    Sujeewa AlwisCybula Ltd

  • 8/8/2019 FaceRec Tutorial

    2/32

  • 8/8/2019 FaceRec Tutorial

    3/32

    Cybula 2004

    Why use face?

    Every one has got a fairly unique face

    Can be captured without user cooperation (passive)

  • 8/8/2019 FaceRec Tutorial

    4/32 Cybula 2004

    Application Modes

    Verification Are you the same person you say you are?

    System captures a new biometric sample and the person submits an ID.

    Yes/no answer indicates authentication result.

    Identification Who are you?

    System captures a new biometric sample. It does a database search and presents

    the top n similar matches may need a human operator to make the final decision.

    Watch-list Are we looking for you

    System captures a new biometric sample. System triggers an alarm only if that

    person is in the database. Similar to identification - but uses an additional threshold

    to identify a hit.

  • 8/8/2019 FaceRec Tutorial

    5/32 Cybula 2004

    Iris

    Advantages

    highly unique (five differentpatterns in even two identical twins)

    Stable after the first year of birth

    Disadvantages

    Need user cooperation

    Difficulties during enrolment

    The most successful technique is based on projecting Iris pattern onto a

    Gabor wavelet (Daugman, 1993). Gabor coefficients represent the biometric

    template - commercialised by Iridian technologies

  • 8/8/2019 FaceRec Tutorial

    6/32 Cybula 2004

    Fingerprints

    Advantages

    Availability of large fingerprint databases

    Disadvantages

    Associated with crime control/investigation

    Need user cooperation

    Need to keep the capture surface clean

    and germ-free not suitable for high-throughput applications

    Represents minutiae points in a map

    Cross match technologies is one of the companies that sell fingerprint recognition

    systems

  • 8/8/2019 FaceRec Tutorial

    7/32 Cybula 2004

    Gait recognition

    Palm print recognition

    Voice recognition

  • 8/8/2019 FaceRec Tutorial

    8/32 Cybula 2004

    Combinations

    Face + Iris (Wang, 2003)

    Face + Ear (Chang, 2003)

    Face + Gait (Shakhnanorvich, 2002)

    Face + Palm print + Fingerprint (Ross, 2001)

    Face + Voice + Lip movement (Frischholz, 2000)

    Face + Voice (Kittler, 1997)

  • 8/8/2019 FaceRec Tutorial

    9/32

    Cybula 2004

    Face Representation 2D vs. 3D

    Advantages

    Can deal with pose

    variations if the cameras

    can capture the full face

    Less sensitive to lighting

    variations

    Better accuracy duringrecognition (Experimental

    results from Notre Dame

    University, Chang et al.

    2003)

    Advantages

    Availability of large 2D

    image collections

    Capture devices are

    currently cheaper

    3D2D

  • 8/8/2019 FaceRec Tutorial

    10/32

    Cybula 2004

    Face Representation 2D vs. 3D (contd.)

    DisadvantagesCameras are still expensive

    Takes time to reconstruct

    models

    Unavailability of large

    collections of 3D data

    (UofY/ Cybula data set, U ofNotre Dame data set)

    DisadvantagesCannot handle pose

    variations

    Sensitive to lightingvariations, shadows etc.

    3D2D

  • 8/8/2019 FaceRec Tutorial

    11/32

    Cybula 2004

    Techniques

    Model basedTechniques

    Active appearance/

    shape models,

    Fitting morphable models

    Appearance basedtechniques

    Eigen faces

    and Fisher faces

    Feature basedTechniques

    Distances between landmark

    points such as eyes,

    nose and mouth.

    Graph matching techniques

  • 8/8/2019 FaceRec Tutorial

    12/32

    Cybula 2004

    Eigen Analysis

    One of the most popular methods for face recognition

    The central argument is

    faces contain a lot of features some are common to all faces, some are

    highly discriminatory information.

    So they have to be mapped to different feature space that consists of

    discriminatory information a dimensionality reduction method is needed

    Eigen analysis provides a way to identify dimensions that indicate high

    variance - so we can use Eigen analysis to extract principal components

  • 8/8/2019 FaceRec Tutorial

    13/32

    Cybula 2004

    A simple example

    y = P x

    where

    y coordinates in the new

    space

    x coordinates in the

    previous space

    P projection matrix

    - a face

  • 8/8/2019 FaceRec Tutorial

    14/32

    Cybula 2004

    Eigen Faces

    projections of a face template along different principle components

  • 8/8/2019 FaceRec Tutorial

    15/32

    Cybula 2004

    Previous Work

    Using 2D images

    Sirovich and Kirby (1987), Turk and Pentland (1991) Using 3D images

    Heseltine, Pears and Austin (2003), Chang, Bowyer and Flynn (2003)

  • 8/8/2019 FaceRec Tutorial

    16/32

    Cybula 2004

    Linear Discriminant Analysis

    The aim is to minimise the

    within class separation and

    maximise between class

    separation. In other words,

    maximise the ratio between

    between class variance andwithin class variance

    Maximise (SBSw-1

    )Where

    SB between class scatter matrix

    Sw within class scatter matrix

    Subject A Subject B

    Subject C

  • 8/8/2019 FaceRec Tutorial

    17/32

    Cybula 2004

    Previous Work

    Using 2D images

    Belhumeur, Hespanha and Kriegman (1997), Etemad and Chellappa(1996), Liu and Wiechsler (1998), Kittler (1999)

    Using 3D images

    Heseltine, Pears and Austin (2004)

  • 8/8/2019 FaceRec Tutorial

    18/32

    Cybula 2004

    Is LDA always better than PCA?

    PCA

    LDA

    D PCA

    D LDA Martinez and Kak

    (IEEE PAMI, 2001)Present experimental

    data to show that

    LDA does not always

    outperform

    PCA particularly when

    the number of samples

    in a class is small

  • 8/8/2019 FaceRec Tutorial

    19/32

    Cybula 2004

    Feature based matching techniques

    One of the earliest techniques is to use distance between landmarks such

    as eye, nose and mouth

    This technique may not be robust due to pose variations and it may be

    difficult to accurately identifying the required feature points

  • 8/8/2019 FaceRec Tutorial

    20/32

    Cybula 2004

    Cybula approach

    3D graph matching

    A 3D mesh is used to identify a set of significant points

    we identify high curvature points on face profiles

    These points and the relationships between points are

    represented in a graph

    A graph matching framework called Relaxation by Elimination (RBE)developed at York is used.

  • 8/8/2019 FaceRec Tutorial

    21/32

    Cybula 2004

    Elastic Bunch Graph Matching

    But we are not the only people who have applied graph matching to faces!

    Wiskott, Fellous, Kruger and Malsburg (1999) have used graph matching

    for 2D face recognition.

    Each landmark point (eyes, mouth et.) is represented by a stack of waveletresponses. They become the nodes of the graph. Distances are

    represented in edges.

    Graph for a new image can be fitted by scaling, rotating and translating astandard model graph.

    Dissimilarity measure is a straight-forward comparison between graphs

  • 8/8/2019 FaceRec Tutorial

    22/32

    Cybula 2004

    Model based recognition

    Active appearance models (Cootes, Edwards and Taylor, 2001)

    A statistical appearance model is constructed by combining a shape model

    and a texture model.

    Shape model is constructed by identifying the positions of landmark points

    Texture model represent gray level intensities.

    Model parameters are identified by applying Eigen analysis.

    Recognition is an iterative process in which model parameters are adjusted

    to obtain the best match

  • 8/8/2019 FaceRec Tutorial

    23/32

    Cybula 2004

    3D morphable model (Blanz and Vetter, 2003)

    A set of laser scanned 3D image models (100 males and 100 females) are used to

    construct the morphable 3D model. Shape is represented by 3D co-ordinates while

    texture is represented by colour.

    Model parameters are calculated by applying Eigen analysis.

    3D model is deformed to obtain the best fit between its 2D projection and the new2D image. New model parameters are used to describe the new image. So this couldbe seen as 2D to 3D mapping

    Optimisation process involves finding out optimum values for model parameters aswell as scene parameters (pose, focal length of the camera, light intensity, colour anddirection)

  • 8/8/2019 FaceRec Tutorial

    24/32

    Cybula 2004

    One remaining issue how to keep the data collections

    updated?

    Face is changed when people become older and it could depend on both

    internal and external factors

    Lanitis, Taylor and Cootes (2002) have extended their work on active

    appearance model to predict the age of an unseen subject and then to

    simulate/ eliminate age effects

    Using training data, they build up a weighted person specific aging function

    to predict an age of a person using appearance as well as external factors

    such as lifestyle

    Age simulation can be done by changing the model parameters.

  • 8/8/2019 FaceRec Tutorial

    25/32

    Cybula 2004

    Evaluation

    False acceptance rate (FAR) number of times a wrong person is accepted

    False rejection rate (FRR) - number of times the correct person is rejected

    Equal error rates the value that FAR and FRR becomes equal

    Time to verify

    Time to capture/ enrol

  • 8/8/2019 FaceRec Tutorial

    26/32

    Cybula 2004

    Benchmark Assessments

    FRVT has been replaced by the Grand Challenge Experiment led by NIST

    First round was finished in this month the second round resultssubmission is due next year

  • 8/8/2019 FaceRec Tutorial

    27/32

    Cybula 2004

  • 8/8/2019 FaceRec Tutorial

    28/32

    Cybula 2004

  • 8/8/2019 FaceRec Tutorial

    29/32

    Cybula 2004

  • 8/8/2019 FaceRec Tutorial

    30/32

    Cybula 2004

  • 8/8/2019 FaceRec Tutorial

    31/32

    Cybula 2004

  • 8/8/2019 FaceRec Tutorial

    32/32

    Cybula 2004