DOI : 10.3745/JIPS.2009.5.2.041
Invited Paper
Journal of Information Processing Systems, Vol.5, No.2, June 2009 41
A Survey of Face Recognition Techniques
Rabia Jafri* and Hamid R. Arabnia*
Abstract: Face recognition presents a challenging problem in the field of image analysis and
computer vision, and as such has received a great deal of attention over the last few years because of
its many applications in various domains. Face recognition techniques can be broadly divided into
three categories based on the face data acquisition methodology: methods that operate on intensity
images; those that deal with video sequences; and those that require other sensory data such as 3D
information or infra-red imagery. In this paper, an overview of some of the well-known methods in
each of these categories is provided and some of the benefits and drawbacks of the schemes mentioned
therein are examined. Furthermore, a discussion outlining the incentive for using face recognition, the
applications of this technology, and some of the difficulties plaguing current systems with regard to
this task has also been provided. This paper also mentions some of the most recent algorithms
developed for this purpose and attempts to give an idea of the state of the art of face recognition
technology.
Keywords: Face Recognition, Person Identification, Biometrics
1. Problem Definition
The face recognition problem can be formulated as
follows: Given an input face image and a database of face
images of known individuals, how can we verify or
determine the identity of the person in the input image?
2. Why Use the Face for Recognition
Biometric-based techniques have emerged as the most
promising option for recognizing individuals in recent
years since, instead of authenticating people and granting
them access to physical and virtual domains based on
passwords, PINs, smart cards, plastic cards, tokens, keys
and so forth, these methods examine an individual’s
physiological and/or behavioral characteristics in order to
determine and/or ascertain his identity. Passwords and
PINs are hard to remember and can be stolen or guessed;
cards, tokens, keys and the like can be misplaced, forgotten,
purloined or duplicated; magnetic cards can become corrupted
and unreadable. However, an individual’s biological traits
cannot be misplaced, forgotten, stolen or forged.
Biometric-based technologies include identification
based on physiological characteristics (such as face,
fingerprints, finger geometry, hand geometry, hand veins,
palm, iris, retina, ear and voice) and behavioral traits (such
as gait, signature and keystroke dynamics) [1]. Face
Manuscript received 10 March, 2009; accepted 22 April, 2009.
Corresponding Author: Hamid R. Arabnia
* Dept. of Computer Science, University of Georgia, Athens, Georgia,
U.S.A. ({jafri, hra}@cs.uga.edu)
recognition appears to offer several advantages over other
biometric methods, a few of which are outlined here:
Almost all these technologies require some voluntary
action by the user, i.e., the user needs to place his hand on a
hand-rest for fingerprinting or hand geometry detection and
has to stand in a fixed position in front of a camera for iris
or retina identification. However, face recognition can be
done passively without any explicit action or participation
on the part of the user since face images can be acquired
from a distance by a camera. This is particularly beneficial
for security and surveillance purposes. Furthermore, data
acquisition in general is fraught with problems for other
biometrics: techniques that rely on hands and fingers can
be rendered useless if the epidermis tissue is damaged in
some way (i.e., bruised or cracked). Iris and retina
identification require expensive equipment and are much
too sensitive to any body motion. Voice recognition is
susceptible to background noises in public places and
auditory fluctuations on a phone line or tape recording.
Signatures can be modified or forged. However, facial
images can be easily obtained with a couple of inexpensive
fixed cameras. Good face recognition algorithms and
appropriate preprocessing of the images can compensate
for noise and slight variations in orientation, scale and
illumination. Finally, technologies that require multiple
individuals to use the same equipment to capture their
biological characteristics potentially expose the user to the
transmission of germs and impurities from other users.
However, face recognition is totally non-intrusive and does
not carry any such health risks.
Copyright ⓒ 2009 KIPS (ISSN 1976-913X)
42
A Survey of Face Recognition Techniques
3. Applications
Face recognition is used for two primary tasks:
1. Verification (one-to-one matching): When presented
with a face image of an unknown individual along
with a claim of identity, ascertaining whether the
individual is who he/she claims to be.
2. Identification (one-to-many matching): Given an image
of an unknown individual, determining that person’s
identity by comparing (possibly after encoding) that
image with a database of (possibly encoded) images
of known individuals.
There are numerous application areas in which face
recognition can be exploited for these two purposes, a few
of which are outlined below.
• Security (access control to buildings, airports/seaports,
ATM machines and border checkpoints [2, 3]; computer/
network security [4]; email authentication on multimedia
workstations).
• Surveillance (a large number of CCTVs can be
monitored to look for known criminals, drug offenders,
etc. and authorities can be notified when one is located;
for example, this procedure was used at the Super
Bowl 2001 game at Tampa, Florida [5]; in another
instance, according to a CNN report, two cameras
linked to state and national databases of sex offenders,
missing children and alleged abductors have been
installed recently at Royal Palm Middle School in
Phoenix, Arizona [6]).
• General identity verification (electoral registration,
banking, electronic commerce, identifying newborns,
national IDs, passports, drivers’ licenses, employee IDs).
• Criminal justice systems (mug-shot/booking systems,
post-event analysis, forensics).
• Image database investigations (searching image databases
of licensed drivers, benefit recipients, missing children,
immigrants and police bookings).
• “Smart Card” applications (in lieu of maintaining a
database of facial images, the face-print can be stored
in a smart card, bar code or magnetic stripe, authentication of which is performed by matching the live
image and the stored template) [7].
• Multi-media environments with adaptive humancomputer interfaces (part of ubiquitous or contextaware systems, behavior monitoring at childcare or
old people’s centers, recognizing a customer and
assessing his needs) [8, 9].
• Video indexing (labeling faces in video) [10, 11].
• Witness face reconstruction [12].
In addition to these applications, the underlying techniques
in the current face recognition technology have also been
modified and used for related applications such as gender
classification [13-15], expression recognition [16, 17] and
facial feature recognition and tracking [18]; each of these
has its utility in various domains: for instance, expression
recognition can be utilized in the field of medicine for
intensive care monitoring [19] while facial feature
recognition and detection can be exploited for tracking a
vehicle driver’s eyes and thus monitoring his fatigue [20],
as well as for stress detection [21].
Face recognition is also being used in conjunction with
other biometrics such as speech, iris, fingerprint, ear and
gait recognition in order to enhance the recognition
performance of these methods [8, 22-34].
4. General Difficulties
Face recognition is a specific and hard case of object
recognition. The difficulty of this problem stems from the
fact that in their most common form (i.e., the frontal view)
faces appear to be roughly alike and the differences
between them are quite subtle. Consequently, frontal face
images form a very dense cluster in image space which
makes it virtually impossible for traditional pattern
recognition techniques to accurately discriminate among
them with a high degree of success [35].
Furthermore, the human face is not a unique, rigid object.
Indeed, there are numerous factors that cause the
appearance of the face to vary. The sources of variation in
the facial appearance can be categorized into two groups:
intrinsic factors and extrinsic ones [36]. A) Intrinsic factors
are due purely to the physical nature of the face and are
independent of the observer. These factors can be further
divided into two classes: intrapersonal and interpersonal
[37]. Intrapersonal factors are responsible for varying the
facial appearance of the same person, some examples being
age, facial expression and facial paraphernalia (facial hair,
glasses, cosmetics, etc.). Interpersonal factors, however,
are responsible for the differences in the facial appearance
of different people, some examples being ethnicity and
gender. B) Extrinsic factors cause the appearance of the
face to alter via the interaction of light with the face and
the observer. These factors include illumination, pose, scale
and imaging parameters (e.g., resolution, focus, imaging,
noise, etc.).
Evaluations of state-of-the-art recognition techniques
conducted during the past several years, such as the
FERET evaluations [7, 38], FRVT 2000 [39], FRVT 2002
[40] and the FAT 2004 [41], have confirmed that age
variations, illumination variations and pose variations are
three major problems plaguing current face recognition
systems [42].
Rabia Jafri and Hamid R. Arabnia
Although most current face recognition systems work
well under constrained conditions (i.e., scenarios in which
at least a few of the factors contributing to the variability
between face images are controlled), the performance of
most of these systems degrades rapidly when they are put
to work under conditions where none of these factors are
regulated [43].
5. Face Recognition Techniques
The method for acquiring face images depends upon the
underlying application. For instance, surveillance applications
may best be served by capturing face images by means of a
video camera while image database investigations may
require static intensity images taken by a standard camera.
Some other applications, such as access to top security
domains, may even necessitate the forgoing of the nonintrusive quality of face recognition by requiring the user
to stand in front of a 3D scanner or an infra-red sensor.
Therefore, depending on the face data acquisition methodology, face recognition techniques can be broadly divided
into three categories: methods that operate on intensity
images, those that deal with video sequences, and those
that require other sensory data such as 3D information or
infra-red imagery. The following discussion sheds some
light on the methods in each category and attempts to give
an idea of some of the benefits and drawbacks of the
schemes mentioned therein in general (for detailed surveys,
please see [44, 45]).
5.1 Face Recognition from Intensity Images
Face recognition methods for intensity images fall into
two main categories: feature-based and holistic [46-48]. An
overview of some of the well-known methods in these
categories is given below.
5.1.1 Featured-based
Feature-based approaches first process the input image
to identify and extract (and measure) distinctive facial
features such as the eyes, mouth, nose, etc., as well as other
fiducial marks, and then compute the geometric relationships
among those facial points, thus reducing the input facial
image to a vector of geometric features. Standard statistical
pattern recognition techniques are then employed to match
faces using these measurements.
Early work carried out on automated face recognition
was mostly based on these techniques. One of the earliest
such attempts was by Kanade [49], who employed simple
image processing methods to extract a vector of 16 facial
parameters - which were ratios of distances, areas and
43
angles (to compensate for the varying size of the pictures) and used a simple Euclidean distance measure for
matching to achieve a peak performance of 75% on a
database of 20 different people using 2 images per person
(one for reference and one for testing).
Brunelli and Poggio [46], building upon Kanade’s approach,
computed a vector of 35 geometric features (Fig. 1) from a
database of 47 people (4 images per person) and reported a
90% recognition rate. However, they also reported 100%
recognition accuracy for the same database using a simple
template-matching approach.
More sophisticated feature extraction techniques involve
deformable templates ([50], [51], [52]), Hough transform
methods [53], Reisfeld's symmetry operator [54] and Graf's
filtering and morphological operations [55]. However, all
of these techniques rely heavily on heuristics such as
restricting the search subspace with geometrical constraints
[56]). Furthermore, a certain tolerance must be given to the
models since they can never perfectly fit the structures in
the image. However, the use of a large tolerance value
tends to destroy the precision required to recognize
individuals on the basis of the model's final best-fit
parameters and makes these techniques insensitive to the
minute variations needed for recognition [37]. More
recently, Cox et al. [57] reported a recognition performance
of 95% on a database of 685 images (a single image for
each individual) using a 30-dimensional feature vector
derived from 35 facial features (Fig. 2). However, the
facial features were manually extracted, so it is reasonable
to assume that the recognition performance would have
been much lower if an automated, and hence less precise,
feature extraction method had been adopted. In general,
current algorithms for automatic feature extraction do not
provide a high degree of accuracy and require considerable
computational capacity [58].
Fig. 1. Geometrical features (white) used in the face
recognition experiments [46]. (©1993 IEEE)
44
A Survey of Face Recognition Techniques
Fig. 3. Grids for face recognition [61]. (©1999 IEEE)
Fig. 2. 35 manually identified facial features [57]. (©1996
IEEE)
Another well-known feature-based approach is the
elastic bunch graph matching method proposed by
Wiskott et al. [59] . This technique is based on Dynamic
Link Structures [60]. A graph for an individual face is
generated as follows: a set of fiducial points on the face are
chosen. Each fiducial point is a node of a full connected
graph, and is labeled with the Gabor filters’ responses
applied to a window around the fiducial point. Each arch is
labeled with the distance between the correspondent
fiducial points. A representative set of such graphs is
combined into a stack-like structure, called a face bunch
graph. Once the system has a face bunch graph, graphs for
new face images can then be generated automatically by
Elastic Bunch Graph Matching. Recognition of a new face
image is performed by comparing its image graph to those
of all the known face images and picking the one with the
highest similarity value. Using this architecture, the
recognition rate can reach 98% for the first rank and 99%
for the first 10 ranks using a gallery of 250 individuals. The
system has been enhanced to allow it to deal with different
poses (Fig. 3) [61] but the recognition performance on
faces of the same orientation remains the same. Though
this method was among the best performing ones in the
most recent FERET evaluation [62, 63], it does suffer from
the serious drawback of requiring the graph placement
for the first 70 faces to be done manually before the elastic
graph matching becomes adequately dependable [64].
Campadelli and Lanzarotti [65] have recently experimented
with this technique, where they have eliminated the need to
do the graph placement manually by using parametric
models, based on the deformable templates proposed in [50],
to automatically locate fiducial points. They claim to have
obtained the same performances as the elastic bunch graph
employed in [59]. Other recent variations of this approach
replace the Gabor features by a graph matching strategy
[66] and HOGs (Histograms of Oriented Gradients [67].
Considerable effort has also been devoted to recognizing
faces from their profiles [68-72] since, in this case, feature
extraction becomes a somewhat simpler one-dimensional
problem [57, 71]. Kaufman and Breeding [70] reported a
recognition rate of 90% using face profiles; however, they
used a database of only 10 individuals. Harmon et al. [68]
obtained recognition accuracies of 96% on a database of
112 individuals, using a 17-dimensional feature vector to
describe face profiles and utilizing a Euclidean distance
measure for matching. More recently, Liposcak and
Loncaric [71] reported a 90% accuracy rate on a database
of 30 individuals, using subspace filtering to derive a 21dimensional feature vector to describe the face profiles and
employing a Euclidean distance measure to match them
(Fig. 4).
Fig. 4. a) The twelve fiducial points of interest for face
recognition; b) Feature vector has 21 components;
ten distances D1-D10 (normalized with /(D4+D5))
and eleven profile arcs A1-A11 (normalized with
/(A5+A6)) [71]. (Courtesy of Z. Liposcak and S.
Loncaric)
Rabia Jafri and Hamid R. Arabnia
5.1.1.1 Advantages and Disadvantages
The main advantage offered by the featured-based
techniques is that since the extraction of the feature points
precedes the analysis done for matching the image to that
of a known individual, such methods are relatively robust
to position variations in the input image [37]. In principle,
feature-based schemes can be made invariant to size,
orientation and/or lighting [57]. Other benefits of these
schemes include the compactness of representation of the
face images and high speed matching [73].
The major disadvantage of these approaches is the
difficulty of automatic feature detection (as discussed
above) and the fact that the implementer of any of these
techniques has to make arbitrary decisions about which
features are important [74]. After all, if the feature set lacks
discrimination ability, no amount of subsequent processing
can compensate for that intrinsic deficiency [57].
5.1.2 Holistic
Holistic approaches attempt to identify faces using
global representations, i.e., descriptions based on the entire
image rather than on local features of the face. These
schemes can be subdivided into two groups: statistical and
AI approaches. An overview of some of the methods in
these categories follows.
5.1.2.1 Statistical
In the simplest version of the holistic approaches, the
image is represented as a 2D array of intensity values and
recognition is performed by direct correlation comparisons
between the input face and all the other faces in the
database. Though this approach has been shown to work
[75] under limited circumstances (i.e., equal illumination,
scale, pose, etc.), it is computationally very expensive and
suffers from the usual shortcomings of straightforward
correlation-based approaches, such as sensitivity to face
orientation, size, variable lighting conditions, background
clutter, and noise [76]. The major hindrance to the directmatching methods’ recognition performance is that they
attempt to perform classification in a space of very high
dimensionality [76]. To counter this curse of dimensionality,
several other schemes have been proposed that employ
statistical dimensionality reduction methods to obtain and
retain the most meaningful feature dimensions before
performing recognition. A few of these are mentioned below.
Sirovich and Kirby [77] were the first to utilize Principal
Components Analysis (PCA) [78, 79] to economically
represent face images. They demonstrated that any
particular face can be efficiently represented along the
eigenpictures coordinate space, and that any face can be
approximately reconstructed by using just a small
collection of eigenpictures and the corresponding projections
45
(‘coefficients’) along each eigenpicture.
Turk and Pentland [80, 81] realized, based on Sirovich
and Kirby’s findings, that projections along eigenpictures
could be used as classification features to recognize faces.
They employed this reasoning to develop a face recognition
system that builds eigenfaces, which correspond to the
eigenvectors associated with the dominant eigenvalues of
the known face (patterns) covariance matrix, and then
recognizes particular faces by comparing their projections
along the eigenfaces to those of the face images of the
known individuals. The eigenfaces define a feature space
that drastically reduces the dimensionality of the original
space, and face identification is carried out in this reduced
space. An example training set, the average face and the
top seven eigenfaces derived from the training images are
shown in (Figs. 5, 6 and 7), respectively. The method was
tested using a database of 2,500 images of 16 people under
all combinations of 3 head orientations, 3 head sizes or
scales, and 3 lighting conditions and various resolutions.
Recognition rates of 96%, 85% and 64% were reported for
lighting, orientation and scale variation. Though the
method appears to be fairly robust to lighting variations, its
performance degrades with scale changes.
The capabilities of Turk and Pentland’s system have
Fig. 5. An example training set [81]. (With kind permission
of MIT Press Journals)
Fig. 6. The average face [81]. (With kind permission of
MIT Press Journals)
46
A Survey of Face Recognition Techniques
Fig. 7. Seven of the eigenfaces calculated from the images
of Fig. 5, without the background removed [81].
(With kind permission of MIT Press Journals)
been extended in several ways in [82] and tested on a
database of 7,562 images of approximately 3,000 people. A
“multiple observer” method has been suggested to deal
with large changes in pose: Given N individuals under M
different views, one can either do recognition and pose
estimation in a universal eigenspace calculated from the
combination of NM images (parametric approach) or,
alternatively, one can build a set of M separate eigenspaces,
one for each of the N views (the view-based approach).
The view-based approach is reported to have yielded better
results than the parametric one. A modular “eigenfeatures”
approach has also been proposed to deal with localized
variations in the facial appearance where a low-resolution
description of the whole face is augmented by additional
higher resolution details in terms of the salient facial
features (Fig. 8). This system is reported to have produced
slightly better results than the basic eigenfaces approach
(Figs. 9, 10). Though no implementation has been reported,
it has however been suggested in [81] that variation in
scale be dealt with by employing multi-scale eigenfaces or
by rescaling the input image to multiple sizes and using the
scale that results in the smallest distance measure to the
face space. PCA appears to work well when a single image
of each individual is available, but when multiple images
per person are present, then Belhumeur et al. [83] argue
that by choosing the projection which maximizes total
scatter, PCA retains unwanted variations due to lighting
and facial expression. As stated by Moses et al. [84], “the
variations between the images of the same face due to
illumination and lighting direction are almost always larger
than image variations due to a change in face identity” (see
Fig. 11 for an example of this.) Therefore, they propose
Fig. 8. (a) Examples of facial feature training templates
used and (b) the resulting typical detections [82].
(©1994 IEEE)
Fig. 9. Recognition rates for eigenfaces, eigenfeatures, and
the combined modular representation [82]. (©1994
IEEE)
using Fisher’s Linear Discriminant Analysis [85], which
maximizes the ratio of the between-class scatter and the
Rabia Jafri and Hamid R. Arabnia
47
(a)
Fig. 10. (a) Test views, (b) Eigenface matches, (c) Eigenfeature matches [82]. (©1994 IEEE)
(b)
Fig. 12. (a) This sample of eigenfaces shows the tendency
of the principal components to capture major
variations in the training set such as lighting
direction; (b) The corresponding sample of Fisherfaces shows the ability of Fisherfaces to discount
those factors unrelated to classification [86].
(©1996 IEEE).
Fig. 11. The same person seen under varying light
conditions can appear dramatically different [83].
(©1997 IEEE)
PCA can outperform LDA and also that PCA is less
sensitive to different training sets.
The standard eigenfaces and the Fisherfaces approaches
assume the existence of an optimal projection that projects
the face images to distinct non-overlapping regions in the
reduced subspace where each of these regions corresponds
to a unique subject. However, in reality, that assumption
may not necessarily be true since images of different
people may frequently map to the same region in the face
space and, thus, the regions corresponding to different
individuals may not always be disjoint.
Moghaddam et al. [88] propose an alternative approach
which utilizes difference images, where a difference image
for two face images is defined as the signed arithmetic
difference in the intensity values of the corresponding
pixels in those images. Two classes of difference images
are defined: intrapersonal, which consists of difference
images originating from two images of the same person,
and extrapersonal, which consists of difference images
derived from two images of different people.
It is assumed that both these classes originate from
discrete Gaussian distributions within the space of all
possible difference images. Then, given the difference
image between two images I1 and I2, the probability that
the difference image belongs to the intrapersonal class is
given by Bayes Rule as follows:
within-class scatter and is thus purportedly better for
classification than PCA. Conducting various tests on 330
images of 5 people (66 of each), they report that their
method, called Fisherfaces, which uses subspace projection
prior to LDA projection (to prevent the within-class scatter
matrix from becoming degenerate), is better at simultaneously handling variations in lighting and expression.
Swets and Weng [86] previously reported similar results
when employing the same procedure not only for faces but
also for general objects (90% accuracy on a database of
1316+298 images from 504 classes) (Fig. 12 shows some
examples of eigenfaces and Fisherfaces and how
Fisherfaces capture discriminatory information better than
eigenfaces). It should be noted, however, that some recent
work [87] shows that when the training data set is small,
P(ΩI | d (I1 , I 2 )) =
P(d (I1 , I 2 ) | ΩI )P(ΩI )
P(d (I1 , I 2 ) | ΩI )P(ΩI ) + P(d (I1 , I 2 ) | ΩE )P(ΩE ) (1)
48
A Survey of Face Recognition Techniques
where
d(I1, I2) = the difference image between two images I1
and I2
ΩI
= the intrapersonal class
ΩE
= the extrapersonal class
The formulation of the face recognition problem in this
manner converts it from an m-ary classification problem
(where m is the number of individuals in the database of
known face images) into a binary classification problem
which can be solved using the maximum a posteriori
(MAP) rule – the two images are declared to belong to the
same individual if P(ΩI|d(I1, I2)) > P(ΩE|d(I1, I2)) or,
equivalently, if P(ΩI|d(I1, I2)) > ½. For a computationally
more expedient approach, Moghaddam and Pentland [89]
also suggest ignoring the extrapersonal class information
and calculating the similarity based only on the
intrapersonal class information. In the resulting maximum
likelihood (ML) classifier, the similarity score is given only
by P(d(I1, I2)| ΩI).
Numerous variations on and extensions to the standard
eigenfaces and the Fisherfaces approaches have been
suggested since their introduction. Some recent advances
in PCA-based algorithms include multi-linear subspace
analysis [90], symmetrical PCA [91], two-dimensional
PCA [92, 93], eigenbands [94], adaptively weighted subpattern PCA [95], weighted modular PCA [96], Kernel
PCA [97, 98] and diagonal PCA [99]. Examples of recent
LDA-based algorithms include Direct LDA [100, 101],
Direct-weighted LDA [102], Nullspace LDA [103, 104],
Dual-space LDA [105], Pair-wise LDA [106], Regularized
Discriminant Analysis [107], Generalized Singular Value
Decomposition [108, 109], Direct Fractional-Step LDA
[110], Boosting LDA [111], Discriminant Local Feature
Analysis [112], Kernel PCA/LDA [113, 114], Kernel
Scatter-Difference-based Discriminant Analysis [115], 2DLDA [116, 117], Fourier-LDA [118], Gabor-LDA [119],
Block LDA [120], Enhanced FLD [121], Component-based
Cascade LDA [122], and incremental LDA [123], to name
but a few. All these methods purportedly obtain better
recognition results than the baseline techniques.
One main drawback of the PCA and LDA methods is
that these techniques effectively see only the Euclidean
structure and fail to discover the underlying structure if the
face images lie on a non-linear submanifold in the image
space. Since it has been shown that face images possibly
reside on a nonlinear submanifold [124-130] (especially if
there is a perceivable variation in viewpoint, illumination
or facial expression), some nonlinear techniques have
consequently been proposed to discover the nonlinear
structure of the manifold, e.g., Isometric Feature Mapping
(ISOMAP) [130], Locally Linear Embedding (LLE) [126,
131], Laplacian Eigenmap [132], Locality Preserving
Projection (LPP) [133], Embedded Manifold [134],
Nearest Manifold Approach [135], Discriminant Manifold
Learning [136] and Laplacianfaces [137].
The eigenvectors found by PCA depend only on pairwise relationships between the pixels in the image database.
However, other methods exist that can find basis vectors
that depend on higher-order relationships among the pixels,
and it seems reasonable to expect that utilizing such
techniques would yield even better recognition results.
Independent component analysis (ICA) [138], a generalization of PCA, is one such method that has been employed
for the face recognition task. ICA aims to find an independent, rather than an uncorrelated, image decomposition
and representation. Bartlett et al. [139] performed ICA on
images in the FERET database under two different
architectures: one treated the images as random variables
and the pixels as outcomes; conversely, the second treated
the pixels as the random variables and the images as
outcomes. Both ICA representations outperformed PCA
representations for recognizing faces across days and
changes in expression. A classifier that combined both ICA
representations gave the best performance. Others have
also experimented with ICA [140-146] and have reported
that this technique, and variations of it, appear to perform
better then PCA under most circumstances.
Other subspace methods have also been exploited for the
face recognition task: Foon et al. [147] have integrated
various wavelet transforms and non-negative matrix
factorizations [148] and claim to have obtained better
verification rates as compared to the basic eigenfaces
approach. In [149], an intra-class subspace is constructed,
and the classification is based on the nearest weighted
distance between the query face and each intra-class
subspace. Experimental results are presented to demonstrate
that this method performs better than some other nearest
feature techniques.
A study and comparison of four subspace representations
for face recognition, i.e., PCA, ICA, Fisher Discriminant
Analysis (FDA), and probabilistic eigenfaces and their
‘kernalized’ versions (if available), is presented in [150]. A
comprehensive review of recent advances in subspace
analysis for face recognition can be found in [151].
5.1.2.2 AI
AI approaches utilize tools such as neural networks and
machine learning techniques to recognize faces. Some
examples of methods belonging to this category are given
below.
In [152], 50 principal components were extracted and an
auto-associative neural network was used to reduce those
components to five dimensions. A standard multi-layer
perceptron was exploited to classify the resulting repre-
Rabia Jafri and Hamid R. Arabnia
sentation. Though favorable results were received, the
database used for training and testing was quite simple: the
pictures were manually aligned, there was no lighting
variation, tilting, or rotation, and there were only 20 people
in the database.
Weng et al. [153] made use of an hierarchical neural
network which was grown automatically and not trained on
the traditional gradient descent method. They reported
good results on a database of 10 subjects.
Lawrence et al [58] reported a 96.2% recognition rate on
the ORL database (a database of 400 images of 40
individuals) using a hybrid neural network solution which
combines local image sampling, a self-organizing map
[154, 155] neural network (which provides dimensionality
reduction and invariance to small changes in the image
sample), and a convolutional neural network (which
provides partial invariance to translation, rotation, scale
and deformation). The eigenfaces method [80, 81]
produced 89.5% recognition accuracy on the same data.
Replacing the self-organizing map by the Karhunen-Loeve
transform and the convolutional network by a multi-layer
perceptron resulted in a recognition rate of 94.7% and 60%
respectively (Fig. 13).
Eleyan and Demirel [156] used principal components
analysis to obtain feature projection vectors from face
images which were then classified using feed forward
neural networks. Some tests on the ORL database using
various numbers of training and testing images showed that
the performance of this system was better than the
eigenfaces [80, 81] one in which a nearest neighbor
classifier was used for classification.
Li and Yin [157] introduced a system in which a face
image is first decomposed with a wavelet transform to
three levels. The Fisherfaces method [83] is then applied to
each of the three low-frequency sub-images. Then, the
individual classifiers are fused using the RBF neural
network. The resulting system was tested on images of 40
subjects from the FERET database and was shown to
outperform the individual classifiers and the direct
Fisherfaces method.
Fig. 13. A high-level diagram of the system used for face
recognition [58]. (©1997 IEEE)
49
Melin et al. [158] divided the face into three regions (the
eyes, the mouth, and the nose) and assigned each region to
a module of the neural network. A fuzzy Sugeno integral
was then used to combine the outputs of the three modules
to make the final face recognition decision. They tested it
on a small database of 20 people and reported that the
modular network yielded better results than a monolithic one.
Recently, Zhang et al. [159] proposed an approach in
which a similarity function is learned describing the level
of confidence that two images belong to the same person,
similar to [88]. The facial features are selected by obtaining
Local Binary Pattern (LBP) [160] histograms of the subregions of the face image and the Chi-square distances
between the corresponding LBP histograms are chosen as
the discriminative features. The AdaBoost learning
algorithm, introduced by Freund and Schapire [161], is
then applied to select the most efficient LBP features as
well as to obtain the similarity function in the form of a
linear combination of LBP feature-based weak learners.
Experimental results on the FERET frontal image sets have
shown that this method yields a better recognition rate of
97.9 % by utilizing fewer features than a previous similar
approach proposed by Ahonen et al. [162].
Some researchers have also used the one-against-one
approach [163] for decomposing the multi-class face
recognition problem into a number of binary classification
problems. In this method, one classifier is trained for each
pair of classes, ignoring all the remaining ones. The
outputs of all the binary classifiers are then combined to
construct the global result. For binary classifiers with
probabilistic outputs, pair-wise coupling (PWC) [164] can
be used to couple these outputs into a set of posterior
probabilities. Then, the test example is assigned to the class
with the maximum posterior probability. One main disadvantage of PWC is that when a test example does not
belong to either of the classes related to a binary classifier,
then the output of that classifier is meaningless and can
damage the global result. In [165], a new algorithm called
PWC-CC (where CC stands for correcting classifier) is
presented to solve this problem: for each binary classifier
separating class ci from class cj, a new classifier separating
the two classes from all other classes is trained. Even
though PWC-CC performs better than PWC, it has its own
drawbacks. In [166], a novel PWC-CC (NPWC-CC)
method is proposed for the face recognition problem and
the results of tests on the ORL database are presented to
support the claim that it outperforms PWC-CC. In [167],
the optimal PWC (O-PWC) approach is introduced and is
shown to have better recognition rates that the PWC
method. Feature extraction is done by using principal
components analysis in [166] and by wavelet transform in
[167]. In both [166] and [167], Support Vector Machines
50
A Survey of Face Recognition Techniques
(SVMs) were employed as binary classifiers and the SVM
outputs were mapped to probabilities by using the method
suggested by Platt [168].
It should be noted that Support Vector Machine (SVM)
is considered to be one of the most effective algorithms for
pattern classification problems [169]. In general, it works
as follows for binary problems [170]: First, the training
examples are mapped to a high-dimensional feature space
H. Then, the optimal hyperplane in H is sought to separate
examples of different classes as much as possible, while
maximizing the distance from either class to the
hyperplane. SVM has been employed for face recognition
by several other researchers and has been shown to yield
good results [169, 171-175].
Hidden Markov models [176] have also been employed
for the face recognition task. Samaria and Harter [177]
used a one-dimensional HMM to obtain a peak recognition
accuracy of 87% on the ORL database. They later upgraded
the one-dimensional HMM to a pseudo two-dimensional
HMM [178] and achieved a best recognition performance
of 95% on the same database using half the images for
training and the other half for testing. Nefian and Hayes III
[179] reported a best recognition rate of 98% on the same
training and testing sets using embedded HMM [180] face
models, and they also claimed that their system was much
faster than that of Samaria [178] and invariant to the scale
of the face images.
Some other AI approaches utilized for the face
recognition task include evolutionary pursuit [181, 182]
and techniques [183, 184] based on boosting [161, 185].
These schemes have reportedly yielded promising results
for various difficult face recognition scenarios.
5.1.2.3 Multiple Classifier Systems
Since the performance of any classifier is more sensitive
to some factors and relatively invariant to others, a recent
trend has been to combine individual classifiers in order to
integrate their complementary information and thereby
create a system that is more robust than any individual
classifier to variables that complicate the recognition task.
Such systems have been termed as multiple classifier
systems (MCSs) [186] and are a very active research area
at present. Examples of such approaches proposed for face
recognition include the following: Lu et al. [187] fused the
results of PCA, ICA and LDA using the sum rule and RBF
network-based [188] integration strategy (Fig. 14);
Marcialis and Roli [189-191] combined the results of the
PCA and LDA algorithms (Fig. 15); Achermann and Bunke
[192] utilized simple fusion rules (majority voting, rank
sum, Baye’s combination rule) to integrate the weighted
outcomes of three classifiers based on frontal and profile
views of faces; Tolba and Abu-Rezq [193] employed a
Fig. 14. Classifier combination system framework [187].
(©2003 IEEE)
Fig. 15. Overview of the fusion methodology [191]. (With
kind permission of Springer Science and Business
Media)
simple combination rule for fusing the decisions of RBF
and LVQ networks; Wan et al. [194] used a SVM and
HMM hybrid model; Kwak and Pedrycz [195] divided the
face into three regions, applied the Fisherfaces method to
the regions as well as to the whole face and then integrated
the classification results using the Choquet fuzzy integral
[196]; Haddadnia et. al. [197] used PCA, the Pseudo
Zernike Moment Invariant (PZMI) [198, 199] and the
Zernike Moment Invariant (ZMI) to extract feature vectors
in parallel, which were then classified simultaneously by
separate RBF neural networks and the outputs of these
networks were then combined by a majority rule to
determine the final identity of the individual in the input
image.
5.1.2.4 Advantages and Disadvantages
The main advantage of the holistic approaches is that
they do not destroy any of the information in the images by
concentrating on only limited regions or points of interest
[37]. However, as mentioned above, this same property is
their greatest drawback, too, since most of these
approaches start out with the basic assumption that all the
pixels in the image are equally important [74]. Consequently, these techniques are not only computationally
expensive but require a high degree of correlation between
the test and training images, and do not perform effectively
under large variations in pose, scale and illumination, etc.
[200]. Nevertheless, as mentioned in the above review,
several of these algorithms have been modified and/or
enhanced to compensate for such variations, and dimen-
Rabia Jafri and Hamid R. Arabnia
51
sionality reduction techniques have been exploited (note
that even though such techniques increase generalization
capabilities, the downside is that they may potentially
cause the loss of discriminative information [151]), as a
result of which these approaches appear to produce better
recognition results than the feature-based ones in general.
In the latest comprehensive FERET evaluation [62, 63], the
probabilistic eigenface [88], the Fisherface [83] and the
EBGM [59] methods were ranked as the best three
techniques for face recognition (Even though the EBGM
method is feature-based in general, its success depends on
its application of holistic neural network methods at the
feature level).
5.2 Face Recognition from Video Sequences
Since one of the major applications of face recognition is
surveillance for security purposes, which involves realtime recognition of faces from an image sequence captured
by a video camera, a significant amount of research has
been directed towards this area in recent years.
A video-based face recognition system typically consists
of three modules: one for detecting the face; a second one
for tracking it; and a third one for recognizing it [201].
Most of these systems choose a few good frames and then
apply one of the recognition techniques for intensity
images to those frames in order to identify the individual
[202]. A few of these approaches are briefly described below.
Howell and Buxton [203] employed a two-layer RBF
network [204, 205] for learning/training and used
Difference of Gaussian (DoG) filtering and Gabor wavelet
analysis for the feature representation, while the scheme
from [206] was utilized for face detection and tracking.
Training and testing were done using two types of image
sequences: 8 primary sequences taken in a relatively
constrained environment, and a secondary sequence recorded
in a much more unconstrained atmosphere (Figs. 16, 17).
The image sequences consisted of 62 to 94 frames. The use
of Gabor wavelet analysis for feature representation, as
opposed to DoG filtering, seemed to yield better recognition
results. The recognition accuracies reported varied quite a
Fig. 16. A complete Primary sequence for the class Carla,
after segmentation but before preprocessing [203].
(©1996 IEEE)
Fig. 17. A complete Secondary sequence for the class
Steve, after segmentation but before preprocessing
[203]. (© 1996IEEE)
bit, ranging from 99%, using 278 images for training and
276 for testing, to 67%, using 16 training and 538 testing
images.
de Campos et al. [207] propose a recognition system
which uses skin color modeling [208] to detect the face,
then utilizes GWN [209] to detect prominent facial
landmarks (i.e., the eyes, nose, and mouth) and to track
those features. For each individual frame, eigenfeatures
[82] are then extracted and a feature selection algorithm
[210] is applied over the combination of all the eigenfeatures,
and the best ones are selected to form the feature space. A
couple of classifiers described in [211] are then applied to
identify the individual in the frame and, finally, a superclassifier based on a voting scheme [212] performs the
final classification for the entire video sequence (Figs. 18,
19). Good recognition results (97.7% accuracy) have been
reported using 174 images of the eyes of 29 people (6
images per person).
Biuk and Loncaric [213] take image sequences in which
the pose of the person’s face changes from -90 degrees to
90 degrees (Fig. 20). The sequences are projected into
eigenspace to give a prototype trajectory for each known
individual. During the recognition phase, an unknown face
trajectory is compared with the prototype trajectories to
determine the identity of the individual (Fig. 21). They
tested the system on a database of 28 individuals (11
frames per person) and found that the system yielded
excellent recognition results when all frames and 4 or more
eigenfaces were used, but that the performance decreased
when either parameter was decreased. They, however,
propose a matching method which associates a score to
each trajectory point and then makes the final match based
on the maximum score: They claim that this enhancement
enables their system to achieve better performance when a
smaller number (< 4) of eigenfaces is used.
52
A Survey of Face Recognition Techniques
Fig. 21. Pattern trajectories for two sequences of the same
person represented with the first three principal
components [213]. (©2001 IEEE)
Fig. 18. Overview of the project [207]. (Courtesy of T. E.
de Campos, R. S. Feris and R. M. Cesar Jr.)
Fig. 19. Feature space generation [207]. (Courtesy of T. E.
de Campos, R. S. Feris and R. M. Cesar Jr.)
Fig. 20. Face images taken from different viewing angles
(profile to profile) [213]. (©2001 IEEE)
Recently, some approaches [214, 215] have employed a
video-to-video paradigm in which information from a
sequence of frames from a video segment is combined and
associated with one individual. This notion involves a
temporal analysis of the video sequence and a condensation of the tracking and recognition problems. Such
schemes are still a matter of ongoing research since the
reported experiments were performed without any real
variations in orientation and facial expressions [216].
It is worth mentioning that several schemes have
incorporated information from other modalities in order to
recognize facial images acquired from video clips. For
instance, [217] makes use of stereo information and reports
a recognition accuracy of 90%, while [8] exploits both
audio and video cues as well as 3D information about the
head to achieve a 100% accuracy rate for 26 subjects. (For
more information about recognition based on audio and
video, see the Proceedings of the AVBPA Conferences [218]).
A detailed survey of recent schemes for face recognition
from video sequences is provided in [219].
5.2.1 Advantages and Disadvantages
Dynamic face recognition schemes appear to be at a
disadvantage relative to their static counterparts in general,
since they are usually hampered by one or more of the
following: low quality images (though image quality may
be enhanced by exploiting super-resolution techniques
[220-223]); cluttered backgrounds (which complicate face
detection [224]); the presence of more than one face in the
picture; and a large amount of data to process [71].
Furthermore, the face image may be much smaller than the
size required by most systems employed by the recognition
modules [202].
However, dynamic schemes do have the following
advantages over static techniques: the enormous abundance
of data empowers the system to choose the frame with the
Rabia Jafri and Hamid R. Arabnia
best possible image and discard less satisfactory ones [203].
Video provides temporal continuity [203], so classification
information from several frames can be combined to
improve recognition performance. Moreover, video allows
the tracking of face images such that variations in facial
expressions and poses can be compensated for, resulting in
improved recognition [225]. Dynamic schemes also have
an edge over static ones when it comes to detecting the
face in a scene, since these schemes can use motion to
segment a moving person’s face [71].
5.3 Face Recognition from Other Sensory Inputs
Though the bulk of the research on face recognition has
been focused on identifying individuals from 2D intensity
images, in recent years some attention has nevertheless
been directed towards exploiting other sensing modalities,
such as 3D or range data and infra-red imagery, for this
purpose.
5.3.1 3D Model-based
The main argument in favor of using 3D information for
face recognition appears to be that it allows us to exploit
features based on the shape and the curvature of the face
(such as the shape of the forehead, jaw line, and cheeks)
without being plagued by the variances caused by lighting,
orientation and background clutter that affect 2D systems
[37, 226, 227]. Another argument for the use of depth data
is that “at our current state of technology, it is the most
straightforward way to input or record complex shape
information for machine analysis” [228]. The obvious
drawbacks of such approaches are their complexity and
computational cost [202].
The following techniques are currently being used to
obtain 3D information [226]:
• Scanning systems: Laser face scanners produced by
companies like Cyberware Inc. [229] and 3D Scanners
Ltd. [230] seem to be producing highly accurate
results; however, the cost of these commercial scanning
services is obviously substantial.
• Structured light systems: These systems make use of
the principles of stereo vision to obtain the range data.
Their main advantage is that the only equipment they
require is cameras and some kind of projection system.
The primary drawback with such systems is that they
can experience difficulty in resolving the shape of the
pattern in the camera image.
• Stereo vision systems: These are systems that attempt
to extract 3D information from two or more 2D images
of the same object taken from different angles. They
are limited to objects which will “generate a sufficient
number of image features to allow for conclusive
53
stereo matching. In the case of trying to establish the
shape of a reasonably smooth object, such as the
human face, these systems would be unable to generate
an accurate surface shape. (Smooth surfaces can be
'roughed up' by the projection of a textured pattern onto
the face)” [226].
• Reverse rendering/shape from shading: These techniques
endeavor to construct the shape of an object using
knowledge about illumination and the physical
properties of the object.
Until recently, there appear to have been very few papers
that describe attempts to recognize faces based mainly on
range data or 3D data about the subjects’ faces. However,
lately, there has been a revival of interest in this area and
several new schemes of this sort have been proposed in the
past few years. One of the earliest of such approaches is
described in [228], where the principle curvatures of the
face surface are calculated from range data (Fig. 22), after
which this data - supplemented by a priori information
about the structure of the face - is used to locate the various
facial features (i.e., the nose eyes, forehead, neck, chin,
etc.). The faces are then normalized to a standard position
and re-interpolated onto a regular cylindrical grid. The
volume of space between two normalized surfaces is used
as a similarity measure. The system was tested using the
face images of 8 people (3 images per person). Features
Fig. 22. Principle curvatures for a single face: magnitude
(a) and direction (c) of maximum curvature,
magnitude (b) and direction (d) of minimum
curvature. Umbilic points are marked in (c) and
(d); filled circles are points with a positive index
and open circles are points with a negative index
[228]. (Courtesy of G. Gordon)
54
A Survey of Face Recognition Techniques
were detected adequately for all faces. Recognition rates of
97% and 100% were reported for individual features and
the whole face respectively.
Another approach described in [200] uses profiles
(which have external contours consisting of rather rigid
parts) instead of frontal images, captures 3D data by
triangulation, and then does a 3D comparison of the profile
data (Figs. 23, 24). This method requires a good deal of
user cooperation and restrictions on the background, etc. in
order for it to work.
[37] uses 3D data to normalize the results obtained from
the face detection algorithm to a form more appropriate for
the recognition engine, i.e., in this case, the 3D data is just
being used to supplement rather than supplant existing face
detection and recognition algorithms.
Some examples of more recent face recognition
approaches based on 3D data include the following:
Castellani et al. [231] approximate the range images of
faces obtained by stereoscopic analysis using Multi-level
B-Splines [232], and SVMs are then used to classify the
resulting approximation coefficients. Some other techniques
[227, 233, 234] first project the 3D face data onto a 2D
Fig. 23. (a) Profile image (b) Contour extraction, nose and
eye localization, feature extraction [200]. (Courtesy
of C. Beumier and M. Acheroy)
Fig. 24. 3D comparison by parallel planar cuts [200].
(Courtesy of C. Beumier and M. Acheroy)
intensity image, whereupon the projected 2D images are
processed as standard intensity images. Yet other methods
have been proposed for 3D face recognition based on local
features [235], local and global geometric cues [236],
profiles [237-240], and the rank-based decision fusion of
various shape-based classifiers [241].
Several approaches have also been proposed that
integrate 2D texture and 3D shape information. Such
methods make use of the PCA of intensity images [242244], facial profile intensities [245], Iterative Closest Point
(ICP [246]) [247, 248], Gabor wavelets [249], and Local
Feature Analysis [250], etc. For instance, Wang et al. [249]
extract 3D shape templates from range images and texture
templates from grayscale images of faces, apply PCA
separately to both kinds of templates to reduce them to
lower-dimensional vectors, then concatenate the shape and
texture vectors and, finally, apply SVMs to the resulting
vectors for classification. In general, experiments with such
systems indicate that combining shape and texture
information reduces the misclassification rates of the face
recognizer.
Comprehensive recent surveys of literature on 3D face
recognition can be found in [251] and [252].
5.3.2 Infra-red
Since thermal infra-red imagery of faces is relatively
insensitive to variations in lighting [253], such images can
hence be used as an option for detecting and recognizing
faces. Furthermore, [254] argues that since infra-red facial
images reveal the vein and tissue structure of the face
which is unique to each individual (like a fingerprint),
some of the face recognition techniques for the visible
spectrum should therefore yield favorable results when
applied to these images. However, there exist a multitude
of factors that discourage the exploitation of such images
for the face recognition task, among which figure the
substantial cost of thermal sensors, the low resolution and
high level of noise in the images, the lack of widely
available data sets of infra-red images, the fact of infra-red
radiation being opaque to glass (making it possible to
occlude part of the face by wearing eyeglasses) [255] and,
last but not least, the fact that infra-red images are sensitive
to changes in ambient temperature, wind and metabolic
processes in the subject [256] (Note that in [257], the use
of blood perfusion data is suggested to alleviate the effect
of ambient temperature).
In [254], the eigenface technique [80, 81] was applied to
a database of 288 hand-aligned low-resolution (160x120)
images of 24 subjects taken from 3 viewpoints. The
following recognition rates were reported: 96% for frontal
views, 96% for 45 degrees views, and 100% for profile
views.
Rabia Jafri and Hamid R. Arabnia
Wilder et al. [258] compared the performance of three
face recognition algorithms on a database of visible and
infra-red images of 101 subjects and concluded that the
recognition results for one modality were not significantly
better than those for the other.
Socolinsky et al. [259] tested the eigenfaces [80, 81] and
the ARENA [260] algorithms on a database of visible and
infrared images of 91 distinct subjects (captured under
various illumination conditions, with varying facial
expressions, and with or without glasses using a sensor
capable of imaging both modalities simultaneously [Figs.
25, 26 and 27]), and reported that the infra-red imagery
significantly outperformed the visible one in all the
classification experiments conducted under the various
above-mentioned conditions.
Selinger and Socolinsky [256] used the same database of
91 subjects and tested the performance of four face
recognition algorithms (PCA, LDA, LFA and ICA) under
the afore-mentioned conditions and reached the same
conclusion, although they did concede that the apparent
superiority of the infra-red approach may stem from the
55
fact that their data did not contain sufficiently challenging
situations (i.e., changes in temperature, wind, etc.) for the
infra-red imagery, whereas it did so for the visible images.
Chen et al. [262] collected several datasets of images
(both in infra-red and the visible spectrum) of 240 distinct
subjects under various expressions and lighting conditions
at different times (some of the images were taken in the
same session while others were taken over a period of ten
weeks [Fig. 28]). They studied the effect of temporal
changes in facial appearance on the performance of the
eigenfaces algorithm in both modalities and concluded that
though the recognition accuracy was approximately the
same for images taken in the same session, visible images
Fig. 25. Sample imagery taken from the database. Note
that LWIR images are not radio-metrically
calibrated [259]. (©2001 IEEE)
Fig. 26. First five visible eigenfaces [259]. (© 2001 IEEE)
Fig. 27. First five LWIR eigenfaces [259]. (©2001 IEEE)
Fig. 28. Normalized face images of one subject in visible
and IR across ten weeks [261]. (Reprinted from
Computer Vision and Image Understanding, 99(3),
X. Chen, P. Flynn and K. Bowyer, IR and Visible
Light Face Recognition, 332-358, Copyright
(2005), with permission from Elsevier).
56
A Survey of Face Recognition Techniques
outperformed the infra-red ones when there was a
significant lapse in the time between which the training and
test images were acquired. They attributed the lower
performance of the infra-red imagery to variations in the
thermal patterns of the same subject and the sensitivity of
the infra-red imagery to the manual location of the eyes.
They also found that the FACEIT [263] software performed
better than eigenfaces in both modalities. However, the
combination of the two classifiers using the sum rule [264]
outperformed the individual classifiers as well as the
FACEIT program. Latter experiments conducted by Chen
et al. [261] on a larger set of data reconfirmed these results.
Other approaches have also been proposed recently
which explore the fusion of face images captured under
visible and infrared light spectrum to improve the
performance of face recognition [255, 265-267].
A comprehensive review of recent advances in face
recognition from infra-red imagery may be found in [268].
[3]
[4]
[5]
[6]
[7]
6. Conclusions
Face recognition is a challenging problem in the field of
image analysis and computer vision that has received a
great deal of attention over the last few years because of its
many applications in various domains. Research has been
conducted vigorously in this area for the past four decades
or so, and though huge progress has been made,
encouraging results have been obtained and current face
recognition systems have reached a certain degree of
maturity when operating under constrained conditions;
however, they are far from achieving the ideal of being
able to perform adequately in all the various situations that
are commonly encountered by applications utilizing these
techniques in practical life. The ultimate goal of
researchers in this area is to enable computers to emulate
the human vision system and, as has been aptly pointed out
by Torres [225], “Strong and coordinated effort between
the computer vision, signal processing, and psychophysics
and neurosciences communities is needed” to attain this
objective.
References
[1]
[2]
A. K. Jain, R. Bolle, and S. Pankanti, "Biometrics:
Personal Identification in Networked Security," A. K.
Jain, R. Bolle, and S. Pankanti, Eds.: Kluwer
Academic Publishers, 1999.
K. Kim, "Intelligent Immigration Control System by
Using Passport Recognition and Face Verification,"
in International Symposium on Neural Networks.
[8]
[9]
[10]
[11]
[12]
[13]
Chongqing, China, 2005, pp.147-156.
J. N. K. Liu, M. Wang, and B. Feng, "iBotGuard: an
Internet-based intelligent robot security system using
invariant face recognition against intruder," IEEE
Transactions on Systems Man And Cybernetics Part
C-Applications And Reviews, Vol.35, pp.97-105,
2005.
H. Moon, "Biometrics Person Authentication Using
Projection-Based Face Recognition System in
Verification Scenario," in International Conference
on Bioinformatics and its Applications. Hong Kong,
China, 2004, pp.207-213.
D. McCullagh, "Call It Super Bowl Face Scan 1," in
Wired Magazine, 2001.
CNN, "Education School face scanner to search for
sex offenders." Phoenix, Arizona: The Associated
Press, 2003.
P. J. Phillips, H. Moon, P. J. Rauss, and S. A. Rizvi,
"The FERET Evaluation Methodology for Face
Recognition Algorithms," IEEE Transactions on
Pattern Analysis and Machine Intelligence, Vol.22,
pp.1090-1104, 2000.
T. Choudhry, B. Clarkson, T. Jebara, and A. Pentland,
"Multimodal person recognition using unconstrained
audio and video," in Proceedings, International
Conference on Audio and Video-Based Person
Authentication, 1999, pp.176-181.
S. L. Wijaya, M. Savvides, and B. V. K. V. Kumar,
"Illumination-tolerant face verification of low-bitrate JPEG2000 wavelet images with advanced
correlation filters for handheld devices," Applied
Optics, Vol.44, pp.655-665, 2005.
E. Acosta, L. Torres, A. Albiol, and E. J. Delp, "An
automatic face detection and recognition system for
video indexing applications," in Proceedings of the
IEEE International Conference on Acoustics, Speech
and Signal Processing, Vol.4. Orlando, Florida, 2002,
pp.3644-3647.
J.-H. Lee and W.-Y. Kim, "Video Summarization and
Retrieval System Using Face Recognition and
MPEG-7 Descriptors," in Image and Video Retrieval,
Vol.3115, Lecture Notes in Computer Science:
Springer Berlin / Heidelberg, 2004, pp.179-188.
C. G. Tredoux, Y. Rosenthal, L. d. Costa, and D.
Nunez, "Face reconstruction using a configural,
eigenface-based composite system," in 3rd Biennial
Meeting of the Society for Applied Research in
Memory and Cognition (SARMAC). Boulder, Colorado,
USA, 1999.
K. Balci and V. Atalay, "PCA for Gender Estimation:
Which Eigenvectors Contribute?" in Proceedings of
Rabia Jafri and Hamid R. Arabnia
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
Sixteenth International Conference on Pattern
Recognition, Vol.3. Quebec City, Canada, 2002, pp.
363-366.
B. Moghaddam and M. H. Yang, "Learning Gender
with Support Faces," IEEE Transactions on Pattern
Analysis and Machine Intelligence, Vol.24, pp.707711, 2002.
R. Brunelli and T. Poggio, "HyperBF Networks for
Gender Classification," Proceedings of DARPA
Image Understanding Workshop, pp.311-314, 1992.
A. Colmenarez, B. J. Frey, and T. S. Huang, "A
probabilistic framework for embedded face and
facial expression recognition," in Proceedings of the
IEEE Conference on Computer Vision and Pattern
Recognition, Vol.1. Ft. Collins, CO, USA, 1999, pp.
1592-1597.
Y. Shinohara and N. Otsu, "Facial Expression
Recognition Using Fisher Weight Maps," in Sixth
IEEE International Conference on Automatic Face
and Gesture Recognition, Vol.100, 2004, pp.499-504.
F. Bourel, C. C. Chibelushi, and A. A. Low, "Robust
Facial Feature Tracking," in British Machine Vision
Conference. Bristol, 2000, pp.232-241.
K. Morik, P. Brockhausen, and T. Joachims,
"Combining statistical learning with a knowledgebased approach -- A case study in intensive care
monitoring," in 16th International Conference on
Machine Learning (ICML-99). San Francisco, CA,
USA: Morgan Kaufmann, 1999, pp.268-277.
S. Singh and N. Papanikolopoulos, "Vision-based
detection of driver fatigue," Department of Computer
Science, University of Minnesota, Technical report
1997.
D. N. Metaxas, S. Venkataraman, and C. Vogler,
"Image-Based Stress Recognition Using a ModelBased Dynamic Face Tracking System," International
Conference on Computational Science, pp.813-821,
2004.
M. M. Rahman, R. Hartley, and S. Ishikawa, "A
Passive And Multimodal Biometric System for
Personal Identification," in International Conference
on Visualization, Imaging and Image Processing.
Spain, 2005, pp.89-92.
R. Brunelli and D. Falavigna, "Person identification
using multiple cues," IEEE Transactions on Pattern
Analysis and Machine Intelligence, Vol.17, pp.955966, 1995.
M. Viswanathan, H. S. M. Beigi, A. Tritschler, and F.
Maali, "Information access using speech, speaker
and face recognition," in IEEE International
Conference on Multimedia and Expo, Vol.1, 2000, pp.
57
493--496.
[25] A. K. Jain, K. Nandakumar, X. Lu, and U. Park,
"Integrating Faces, Fingerprints, and Soft Biometric
Traits for User Recognition," Proceedings of Biometric
Authentication Workshop, in conjunction with ECCV2004,
LNCS 3087, pp.259-269, 2004.
[26] P. Melin and O. Castillo, "Human Recognition using
Face, Fingerprint and Voice," in Hybrid Intelligent
Systems for Pattern Recognition Using Soft Computing,
Vol.172, Studies in Fuzziness and Soft Computing:
Springer Berlin / Heidelberg, 2005, pp.241-256.
[27] K. Chang, K. W. Bowyer, S. Sarkar, and B. Victor,
"Comparison and Combination of Ear and Face
Images in Appearance-Based Biometrics," IEEE
Transactions on Pattern Analysis and Machine
Intelligence, Vol.25, pp.1160-1165, 2003.
[28] R. Chellappa, A. Roy-Chowdhury, and S. Zhou,
"Human Identification Using Gait and Face," in The
Electrical Engineering Handbook, 3rd ed: CRC
Press, 2004.
[29] S. Ben-Yacoub, J. Luttin, K. Jonsson, J. Matas, and J.
Kittler, "Audio-visual person verification," in IEEE
Computer Society Conference on Computer Vision
and Pattern Recognition, Vol.1. Fort Collins, CO,
USA, 1999, pp.580-585.
[30] X. Zhou and B. Bhanu, "Feature fusion of side face
and gait for video-based human identification,"
Pattern Recognition, Vol.41, pp.778-795, 2008.
[31] D. Bouchaffra and A. Amira, "Structural hidden Markov
models for biometrics: Fusion of face and fingerprint,"
Pattern Recognition, Vol.41, pp.852-867, 2008.
[32] H. Vajaria, T. Islam, P. Mohanty, S. Sarkar, R. Sankar,
and R. Kasturi, "Evaluation and analysis of a face
and voice outdoor multi-biometric system," Pattern
Recognition Letters, Vol.28, pp.1572-1580, 2007.
[33] Y.-F. Yao, X.-Y. Jing, and H.-S. Wong, "Face and
palmprint feature level fusion for single sample
biometrics recognition," Neurocomputing, Vol.70, pp.
1582-1586, 2007.
[34] J. Zhou, G. Su, C. Jiang, Y. Deng, and C. Li, "A face
and fingerprint identity authentication system based
on multi-route detection," Neurocomputing, Vol.70,
pp.922-931, 2007.
[35] C. Nastar and M. Mitschke, "Real time face
recognition using feature combination," in Third
IEEE International Conference on Automatic Face
and Gesture Recognition. Nara, Japan, 1998, pp.
312-317.
[36] S. Gong, S. J. McKenna, and A. Psarrou., Dynamic
Vision: From Images to Face Recognition: Imperial
College Press (World Scientific Publishing Company),
58
A Survey of Face Recognition Techniques
2000.
[37] T. Jebara, "3D Pose Estimation and Normalization
for Face Recognition," Center for Intelligent
Machines, McGill University, Undergraduate Thesis
May, 1996.
[38] P. J. Phillips, H. Wechsler, J.Huang, and P. J. Rauss,
"The FERET database and evaluation procedure for
face-recognition algorithm," Image and Vision
Computing, Vol.16, pp.295-306, 1998.
[39] D. Blackburn, J. Bone, and P. J. Phillips, "Face
recognition vendor test 2000," Defense Advanced
Research Projects Agency, Arlington, VA, Technical
report A269514, February 16, 2001.
[40] P. J. Phillips, P. Grother, R. J. Micheals, D. M.
Blackburn, E. Tabassi, and J. M. Bone, "Face
Recognition Vendor Test (FRVT 2002)," National
Institute of Standards and Technology, Evaluation
report IR 6965, March, 2003.
[41] K. Messer, J. Kittler, M. Sadeghi, M. Hamouz, A.
Kostin, F. Cardinaux, S. Marcel, S. Bengio, C.
Sanderson, J. Czyz, L. Vandendorpe, C. McCool, S.
Lowther, S. Sridharan, V. Chandran, R. P. Palacios, E.
Vidal, L. Bai, L. Shen, Y. Wang, Y.-H. Chiang, H.-C.
Liu, Y.-P. Hung, A. Heinrichs, M. Müller, A. Tewes,
C. v. d. Malsburg, R. P. Würtz, Z. Wang, F. Xue, Y.
Ma, Q. Yang, C. Fang, X. Ding, S. Lucey, R. Goss, H.
Schneiderman, N. Poh, and Y. Rodriguez, "Face
Authentication Test on the BANCA Database," in
17th International Conference on Pattern Recognition,
Vol.4. Cambridge, UK, 2004, pp.523-532.
[42] X. Q. Ding and C. Fang, "Discussions on some
problems in face recognition," in Advances In
Biometric Person Authentication, Proceedings, Vol.
3338, Lecture Notes In Computer Science: Springer
Berlin / Heidelberg, 2004, pp.47-56.
[43] J. Yang, X. Chen, and W. Kunz, "A PDA-based face
recognition system," in Proceedings of sixth IEEE
Workshop on Applications of Computer Vision.
Orlando, Florida, 2002, pp.19-23.
[44] W. Zhao, R. Chellappa, P. Phillips, and A. Rosenfeld,
"Face Recognition: A Literature Survey," ACM
Computing Surveys, Vol.35, pp.399-458, 2003.
[45] A. F. Abate, M. Nappi, D. Riccio, and G. Sabatino,
"2D and 3D face recognition: A survey," Pattern
Recognition Letters, Vol.28, pp.1885-1906, 2007.
[46] R. Brunelli and T. Poggio, "Face recognition: features
versus templates," IEEE Transactions on Pattern
Analysis and Machine Intelligence, Vol.15, pp.10421052, 1993.
[47] M. A. Grudin, "On internal representations in face
recognition systems," Pattern Recognition, Vol.33,
pp.1161-1177, 2000.
[48] B. Heisele, P. Ho, J. Wu, and T. Poggio, "Face
recognition: component-based versus global approaches,"
Computer Vision and Image Understanding, Vol.91,
pp.6-21, 2003.
[49] T. Kanade, "Picture Processing System by Computer
Complex and Recognition of Human Faces," Kyoto
University, Japan, PhD. Thesis 1973.
[50] A. Yuille, D. Cohen, and P. Hallinan, "Feature
extraction from faces using deformable templates,"
in IEEE Computer Society Conference on Computer
Vision and Templates. San Diego, CA, USA, 1989,
pp.104-109.
[51] N. Roeder and X. Li, "Experiments in analyzing the
accuracy of facial feature detection," Vision Interface
'95, pp.8-16, 1995.
[52] C. Colombo, A. D. Bimbo, and S. D. Magistris,
"Human-computer interaction based on eye movement
tracking," Computer Architectures for Machine
Perception, pp.258-263, 1995.
[53] M. Nixon, "Eye spacing measurement for facial
recognition," in SPIE Proceedings, 1985, pp.279-285.
[54] D. Reisfeld, "Generalized symmetry transforms:
attentional mechanisms and face recognition," TelAviv University, PhD. Thesis, technical report 1994.
[55] H. P. Graf, T. Chen, E. Petajan, and E. Cosatto,
"Locating faces and facial parts," in International
Workshop on Automatic Face- and GestureRecognition, 1995, pp.41-46.
[56] I. Craw, D. Tock, and A. Bennett, "Finding face
features," in Second European Conference on
Computer Vision, 1992, pp.92-96.
[57] I. J. Cox, J. Ghosn, and P. N. Yianilos, "Featurebased face recognition using mixture-distance," in
Proceedings of IEEE Conference on Computer
Vision and Pattern Recognition, 1996, pp.209-216.
[58] S. Lawrence, C. L. Giles, A. C. Tsoi, and A. D. Back,
"Face Recognition: A Convolutional Neural Network
Approach," IEEE Transactions on Neural Networks,
Special Issue on Neural Networks and Pattern
Recognition, pp.1-24, 1997.
[59] L. Wiskott, J.-M. Fellous, N. Krüger, and C. von der
Malsburg, "Face Recognition by Elastic Bunch
Graph Matching," IEEE Transactions on Pattern
Analysis and Machine Intelligence, Vol.19, pp.775779, 1997.
[60] M. Lades, J. C. Vorbrüggen, J. Buhmann, J. Lange, C.
v. d. Malsburg, R. P. Würtz, and W. Konen,
"Distortion invariant object recognition in the
dynamic link architecture," IEEE Trans. Computers,
Vol.42, pp.300-311, 1993.
Rabia Jafri and Hamid R. Arabnia
[61] L. Wiskott, J. M. Fellous, N. Krüger, and C. von der
Malsburg, "Face Recognition by Elastic Bunch
Graph Matching," in Intelligent Biometric Techniques
in Fingerprint and Face Recognition, L. C. Jain, U.
Halici, I. Hayashi, S. B. Lee, and Jae-Ho, Eds.: CRC
Press, 1999, pp.355-396.
[62] P. J. Phillips, P. Rauss, and S. Der, "FERET (FacE
REcognition Technology) Recognition Algorithm
Development and Test Report," U.S. Army Research
Laboratory ARL-TR-995, 1996.
[63] P. J. Phillips, H. Moon, S. A. Rizvi, and P. J. Rauss,
"The FERET Evaluation Methodology for Facerecognition Algorithms," in Proceedings, IEEE Conference on Computer Vision and Pattern Recognition,
1997, pp.137-143.
[64] G. Sukthankar, "Face recognition: a critical look at
biologically-inspired approaches," Carnegie Mellon
University, Pittsburgh, PA, Technical Report: CMURITR-00-04 2000.
[65] P. Campadelli and R. Lanzarotti, "A Face Recognition
System Based on Local Feature Characterization," in
Advanced Studies in Biometrics, Vol.3161, Lecture
Notes in Computer Science, M. Tistarelli, J. Bigun,
and E. Grosso, Eds. Berlin: Springer, 2005, pp.147152.
[66] H. Shin, S. D. Kim, and H. C. Choi, "Generalized
elastic graph matching for face recognition," Pattern
Recognition Letters, Vol.28, pp.1077–1082, 2007.
[67] A. Albiol, D. Monzo, A. Martin, J. Sastre, and A.
Albiol, "Face recognition using HOG–EBGM,"
Pattern Recognition Letters, Vol.29, pp.1537-1543,
2008.
[68] L. D. Harmon, M. K. Khan, R. LAsch, and P. F.
Raming, "Machine Identification of human faces,"
Pattern Recognition, Vol.13, pp.97-110, 1981.
[69] L. D. Harmon, S. C. Kuo, P. F. Raming, and U.
Raudkivi, "Identification of human face profiles by
computers," Pattern Recognition, Vol.10, pp.301-312,
1978.
[70] G. J. Kaufman and K. J. Breeding, "Automatic
recognition of human faces from profile silhouettes,"
IEEE Transactions On Systems Man And Cybernetics,
SMC, Vol.6, pp.113-121, 1976.
[71] Z. Liposcak and S. Loncaric, "A scale-space approach
to face recognition from profiles," in Proceedings of
the 8th International Conference on Computer
Analysis of Images and Patterns, Vol. 1689, Lecture
Notes In Computer Science. London, UK: SpringerVerlag, 1999, pp.243-250.
[72] Z. Liposcak and S. Loncaric, "Face recognition from
profiles using morphological signature transform," in
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
59
Proceedings of the 21st Int'l Conference Information
Technology Interfaces. Pula, Croatia, 1999, pp.93-98.
R. Brunelli and T. Poggio, "Face Recognition Through
Geometrical Features," in Proceedings of the Second
European Conference on Computer Vision, Vol.588,
Lecture Notes In Computer Science, G. Sandini, Ed.
London, UK: Springer-Verlag, 1992, pp.782-800.
R. Cendrillon and B. C. Lowell, "Real-Time Face
Recognition using Eigenfaces," in Proceedings of the
SPIE International Conference on Visual Communications
and Image Processing, Vol.4067, 2000, pp.269-276.
R. J. Baron, "Mechanisms of Human Facial Recognition,"
International Journal of Man-Machine Studies,
Vol.15, pp.137-178, 1981.
R.-J. J. Huang, "Detection Strategies for face
recognition using learning and evolution," George
Mason University, Fairfax, Virginia, Ph. D. Dissertation
1998.
L. Sirovich and M. Kirby, "Low-dimensional Procedure
for the Characterization of Human Faces," Journal of
the Optical Society of America A: Optics, Image
Science, and Vision, Vol.4, pp.519-524, 1987.
A. K. Jain and R. C. Dubes, Algorithms for Clustering
Data. New Jersey: Prentice-Hall, 1988.
K. Fukunaga, Introduction to Statistical Pattern
Recognition, second ed. Boston, MA: Academic
Press, 1990.
M. Turk and A. Pentland, "Face Recognition Using
Eigenfaces," in Proceedings of the IEEE Conference
on Computer Vision and Pattern Recognition, 1991,
pp.586-591.
M. Turk and A. Pentland, "Eigenfaces For Recognition,"
Journal Of Cognitive Neuroscience, Vol.3, pp.71-86,
1991.
A. Pentland, B. Moghaddam, and T. Starner, "Viewbased and modular eigenspaces for face recognition,"
in IEEE Conference on Computer Vision and Pattern
Recognition, 1994, pp.84-90.
P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman,
"Eigenfaces vs. Fisherfaces: Recognition using class
specific linear projection," IEEE Transactions on
Pattern Analysis and Machine Intelligence, Vol.19,
pp.711-720, 1997.
Y. Moses, Y. Adini, and S. Ullman, "Face recognition:
the problem of compensating for changes in
illumination direction," in European Conf. Computer
Vision, 1994, pp.286-296.
R. A. Fisher, "The use of multiple measures in
taxonomic problems," Annals of Eugenics, Vol.7, pp.
179-188, 1936.
D. L. Swets and J. J. Weng, "Using discriminant
60
[87]
[88]
[89]
[90]
[91]
[92]
[93]
[94]
[95]
[96]
[97]
[98]
A Survey of Face Recognition Techniques
eigenfeatures for image retrieval," IEEE Transactions
On Pattern Analysis And Machine Intelligence, Vol.
18, pp.831-836, 1996.
A. M. Martínez and A. C. Kak, "PCA versus LDA,"
IEEE Transactions on Pattern Analysis and Machine
Intelligence, Vol.23, pp.228-233, 2001.
B. Moghaddam, C. Nastar, and A. Pentland, "A
Bayesian Similarity Measure for Direct Image
Matching," in Proceedings 13th International
Conference on Pattern Recognition, 1996, pp.350358.
B. Moghaddam and A. Pentland, "Probabilistic
visual learning for object representation," IEEE
Transactions On Pattern Analysis And Machine
Intelligence, Vol.19, pp.696-710, 1997.
M. A. O. Vasilescu and D. Terzopoulos, "Multilinear
Subspace Analysis of Image Ensembles," in Proc.
IEEE Int’l Conf. on Computer Vision and Pattern
Recognition, 2003, pp.93-99.
Q. Yang and X. Q. Ding, "Symmetrical Principal
Component Analysis and Its Application in Face
Recognition," Chinese Journal of Computers, Vol.26,
pp.1146–1151, 2003.
J. Yang and D. Zhang, "Two-Dimensional PCA: A
New Approach to Appearance-Based Face Representation and Recognition," IEEE Trans. Pattern
Analysis and Machine Intelligence, Vol.28, pp.131137, 2004.
J. Meng and W. Zhang, "Volume measure in 2DPCAbased face recognition," Pattern Recognition Letters,
Vol.28, pp.1203-1208, 2007.
G. D. C. Cavalcanti and E. C. B. C. Filho, "Eigenbands Fusion for Frontal Face Recognition," in
Proceedings of IEEE Internationall Conference on
Image Processing, Vol.1, 2003, pp.665–668.
K. R. Tan and S. C. Chen, "Adaptively weighted subpattern PCA for face recognition," Neurocomputing,
Vol.64, pp.505-511, 2005.
A. P. Kumar, S. Das, and V. Kamakoti, "Face
recognition using weighted modular principle component analysis," in Neural Information Processing,
Vol.3316, Lecture Notes In Computer Science: Springer
Berlin / Heidelberg, 2004, pp.362-367.
V. D. M. Nhat and S. Lee, "An Improvement on PCA
Algorithm for Face Recognition," in Advances in
Neural Networks - ISNN 2005, Vol.3498, Lecture
Notes in Computer Science. Chongqing: Springer,
2005, pp.1016-1021.
N. Sun, H.-x. Wang, Z.-h. Ji, C.-r. Zou, and L. Zhao,
"An efficient algorithm for Kernel two-dimensional
principal component analysis," Neural Computing &
Applications, Vol.17, pp.59-64, 2008.
[99] D. Zhang, Z.-H. Zhoua, and S. Chen, "Diagonal
principal component analysis for face recognition,"
Pattern Recognition, Vol.39, pp.140-142, 2006.
[100] H. Yu and J. Yang, "A Direct LDA Algorithm for
High-dimensional Data with Application to Face
Recognition," Pattern Recognition, Vol.34, pp.20672070, 2001.
[101] F. Song, D. Zhang, J. Wang, H. Liu, and Q. Tao, "A
parameterized direct LDA and its application to face
recognition," Neurocomputing, Vol.71, pp.191-196,
2007.
[102] D. Zhou and X. Yang, "Face Recognition Using
Direct-Weighted LDA," in 8th Pacific Rim International Conference on Artificial Intelligence.
Auckland, New Zealand, 2004, pp.760-768.
[103] L. Chen, H. Liao, M. Ko, L. J., and G. Yu, " A New
LDA-based Face Recognition System Which Can
Solve the Small Samples Size Problem," Journal of
Pattern Recognition, Vol.33, pp.1713–1726, 2000.
[104] W. Liu, Y. Wang, S. Z. Li, and T. Tan, "Null Space
Approach of Fisher Discriminant Analysis for Face
Recognition," in Biometric Authentication, Vol.3087,
Lecture Notes in Computer Science: Springer Berlin /
Heidelberg, 2004, pp.32-44.
[105] X. Wang and X. Tang, "Dual-space Linear Discriminant Analysis for Face Recognition," in Proceedings
of IEEE International Conference on Computer
Vision and Pattern Recognition, 2004, pp.564–569.
[106] M. Loog, R. P. W. Duin, and R. Haeb-Umbach,
"Multiclass Linear Dimension Reduction by Weighted
Pairwise Fisher Criteria," IEEE Transactions on
Pattern Analysis and Machine Intelligence, Vol.23,
pp.762-766, 2001.
[107] J. H. Friedman, "Regularized Discriminant Analysis,"
Journal of the American Statistical Association,
Vol.84, pp.165-175, 1989.
[108] P. Howland and H. Park, "Generalized Discriminant
Analysis Using the Generalized Singular Value Decomposition," IEEE Trans. On Pattern Analysis and
Machine Intelligence, Vol.26, pp.995–1006, 2004.
[109] J. P. Ye, R. Janardan, C. H. Park, and H. Park, "An
Optimization Criterion for Generalized Discriminant
Analysis on Undersampled Problems," IEEE Trans.
On Pattern Analysis and Machine Intelligence,
Vol.26, pp.982–994, 2004.
[110] J. W. Lu, K. N. Plataniotis, and A. N. Venetsanopoulos,
"Face Recognition Using LDA-based Algorithms,"
IEEE Trans. On Neural Networks, Vol.14, pp.195200, 2003.
[111] J. W. Lu, K. N. Plataniotis, and A. N. Venetsanopoulos,
Rabia Jafri and Hamid R. Arabnia
"Boosting Linear Discriminant Analysis for Face
Recognition," in Proceedings of IEEE International
Conference on Image Processing, Vol.1, 2003,
pp.657-660.
[112] Q. Yang and X. Q. Ding, "Discriminant Local Feature
Analysis of Facial Images," in IEEE International
Conference on Image Processing, Vol.2, 2003,
pp.863-866.
[113] Q. Liu, H. Lu, and S. Ma, "Improving Kernel Fisher
Discriminant Analysis for Face Recognition," IEEE
Transactions on Circuits and Systems for Video
Technology, Vol.14, pp.42-49, 2004.
[114] B. Schölkopf, "Nonlinear Component Analysis as a
Kernel Eigenvalue Problem," Neural Computation,
Vol.10, pp.1299-1319, 1998.
[115] Q. Liu, X. Tang, H. Lu, and S. Ma, "Kernel ScatterDifference Based Discriminant Analysis for Face
Recognition," in Proc. IEEE International Conference
on Pattern Recognition, 2004, pp.419-422.
[116] M. Li and B. Yuan, "2D-LDA: A statistical linear
discriminant analysis for image matrix," Pattern
Recognition Letters, Vol.26, pp.527-532, 2005.
[117] H. L. Xiong, M. N. S. Swamy, and M. O. Ahmad,
"Two-dimensional FLD for face recognition,"
Pattern Recognition, Vol.38, pp.1121-1124, 2005.
[118] X. Y. Jing, Y. Y. Tang, and D. Zhang, "A FourierLDA approach for image recognition," Pattern
Recognition, Vol.38, pp.453-457, 2005.
[119] Y. W. Pang, L. Zhang, M. J. Li, Z. K. Liu, and W. Y.
Ma, "A novel Gabor-LDA based face recognition
method," in Advances In Multimedia Information
Processing - Pcm 2004, Pt 1, Proceedings, vol. 3331,
Lecture Notes In Computer Science, 2004, pp.352358.
[120] V. D. M. Nhat and S. Lee, "Block LDA for Face
Recognition," in Computational Intelligence and
Bioinspired Systems, Vol.3512, Lecture Notes in
Computer Science: Springer Berlin / Heidelberg,
2005, pp.899-905.
[121] D. Zhou and X. Yang, "Face Recognition Using
Enhanced Fisher Linear Discriminant Model with
Facial Combined Feature," in PRICAI 2004: Trends
in Artificial Intelligence, Vol.3157, Lecture Notes in
Computer Science: Springer Berlin / Heidelberg,
2004, pp.769-777.
[122] W. C. Zhang, S. G. Shan, W. Gao, Y. Z. Chang, and B.
Cao, "Component-based cascade linear discriminant
analysis for face recognition," in Advances In
Biometric Person Authentication, Proceedings,
Vol.3338, Lecture Notes In Computer Science, 2004,
pp.288-295.
61
[123] H. Zhao and P. C. Yuen, "Incremental Linear Discriminant
Analysis for Face Recognition," IEEE Transactions
on Systems, Man & Cybernetics: Part B, Vol.38,
pp.210-221, 2008.
[124] Y. Chang, C. Hu, and M. Turk, "Manifold of facial
expression," in IEEE International Workshop on
Analysis and Modeling of Faces and Gestures, 2003,
pp.28-35.
[125] K.-C. Lee, J. Ho, M.-H. Yang, and D. J. Kriegman,
"Video-Based Face Recognition Using Probabilistic
Appearance Manifolds," Computer Vision and Pattern
Recognition, Vol.1, pp.313-320, 2003.
[126] S. T. Roweis and L. K. Saul, "Nonlinear Dimensionality Reduction by Locally Linear Embedding,"
Science, Vol.290, pp.2323–2326, 2000.
[127] S. Roweis, L. Saul, and G. E. Hinton, "Global coordination of local linear models," Advances in Neural
Information Processing Systems, Vol.14, pp.889-896,
2002.
[128] H. Seung and D. Lee, "The Manifold Ways of
Perception," Science, Vol.290, pp.2268-2269, 2000.
[129] A. Shashua, A. Levin, and S. Avidan, "Manifold
Pursuit: A New Approach to Appearance Based
Recognition," in International Conference on Pattern
Recognition, Vol.3, 2002, pp.590-594.
[130] J. Tenenbaum, V. de Silva, and J. Langford, "A
global geometric framework for nonlinear dimensionality reduction," Science, Vol.290, pp.2319-2323,
2000.
[131] L. K. Saul and S. T. Roweis, "Think Globally, Fit
Locally: Unsupervised Learning of Low Dimensional Manifolds," Journal of Machine Learning
Research, Vol.4, pp.119-155, 2003.
[132] M. Belkin and P. Niyogi, "Laplacian eigenmaps and
spectral techniques for embedding and clustering,"
Advances in Neural Information Processing Systems,
Vol.14, pp.585-591, 2001.
[133] X. He, S. C. Yan, Y. X. Hu, and H. J. Zhang,
"Learning a Locality Preserving Subspace for Visual
Recognition," in Proceedings of 9th IEEE International Conference on Computer Vision, Vol. 1,
2003, pp.385-392.
[134] S. C. Yan, H. J. Zhang, Y. X. Hu, B. Y. Zhang, and Q.
S. Cheng, "Discriminant Analysis on Embedded
Manifold," in European Conference on Computer
Vision, Vol. LNCS 3021: Springer Berlin / Heidelberg,
2004, pp.121-132.
[135] J. Zhang, S. Z. Li, and J. Wang, "Nearest Manifold
Approach for Face Recognition," in Proc. IEEE
International Conference on Automatic Face and
Gesture Recognition, 2004, pp.223-228.
62
A Survey of Face Recognition Techniques
[136] Y. Wu, K. L. Chan, and L. Wang, "Face Recognition
based on Discriminative Manifold Learning," in
Proc. IEEE Int’l Conf. on Pattern Recognition, Vol.4,
2004, pp.171-174.
[137] X. He, S.-C. Yan, Y. Hu, P. Niyogi, and H.-J. Zhang,
"Face Recognition Using Laplacianfaces," IEEE
Transactions on Pattern Analysis and Machine
Intelligence, Vol.27, pp.328-340, 2005.
[138] P. Comon, "Independent component analysis—A new
concept?" Signal Processing, Vol.36, pp.287-314, 1994.
[139] M. S. Bartlett, J. R. Movellan, and T. J. Sejnowski,
"Face recognition by independent component analysis,"
IEEE Transactions on Neural Networks, Vol.13,
pp.1450-1464, 2002.
[140] B. Draper, K. Baek, M. S. Bartlett, and J. R.
Beveridge, "Recognizing faces with PCA and ICA,"
Computer Vision and Image Understanding: Special
Issue on Face Recognition, Vol.91, pp.115-137, 2003.
[141] C. Liu and H. Wechsler, "Comparative Assessment
of Independent Component Analysis (ICA) for Face
Recognition," in International Conference on Audio
and Video Based Biometric Person Authentication.
Washington, D.C., 1999, pp.211-216.
[142] J. Kim, J. Choi, and J. Yi, "Face Recognition Based
on Locally Salient ICA Information," in Biometric
Authentication Workshop, vol. 3087, Lecture Notes
in Computer Science: Springer Berlin / Heidelberg,
2004, pp.1-9.
[143] J. Yi, J. Kim, J. Choi, J. Han, and E. Lee, "Face
Recognition Based on ICA Combined with FLD,"
Biometric Authentication, pp.10-18, 2002.
[144] T. Martiriggiano, M. Leo, T. D'Orazio, and A.
Distante, "Face Recognition by Kernel Independent
Component Analysis," in The 18th International
Conference on Industrial & Engineering Applications
of Artificial Intelligence & Expert Systems. Bari,
Italy, 2005, pp.55-58.
[145] J. Kim, J. Choi, and J. Yi, "ICA Based Face
Recognition Robust to Partial Occlusions and Local
Distortions," in International Conference on Bioinformatics and its Applications. Fort Lauderdale,
Florida, USA, 2004, pp.147-154.
[146] K.-C. Kwak and W. Pedrycz, "Face Recognition
Using an Enhanced Independent Component Analysis
Approach," IEEE Transactions on Neural Networks,
Vol.18, pp.530-541, 2007.
[147] N. H. Foon, A. T. B. Jin, and D. N. C. Ling, "Face
recognition using wavelet transform and nonnegative matrix factorization," in Ai 2004: Advances
In Artificial Intelligence, Proceedings, Vol.3339, Lecture
Notes In Artificial Intelligence, 2004, pp.192-202.
[148] D. D. Lee and H. S. Seung, "Learning the Parts of
Objects by Non-Negative Matrix Factorization,"
Nature, Vol.401, pp.788-791, 1999.
[149] W. Liu, Y. Wang, S. Z. Li, and T. Tan, "Nearest IntraClass Space Classifier for Face Recognition," in The
17th International Conference on Pattern Recognition
(ICPR), Vol.4. Cambridge, UK, 2004, pp.495-498.
[150] J. Li, S. Zhou, and C. Shekhar, "A Comparison of
Subspace Analysis for Face Recognition," in Proc.
IEEE Int’l Conf. on Acoustics, Speech, and Signal
Processing, 2003, pp.121–124.
[151] Q. Yang and X. Tang, "Recent Advances in Subspace
Analysis for Face Recognition," SINOBIOMETRICS,
pp.275-287, 2004.
[152] D. DeMers and G. W. Cottrell, "Non-linear dimensionality reduction," Advances in Neural Information
Processing Systems, Vol.5, pp.580-587, 1993.
[153] J. Weng, N. Ahuja, and T. S. Huang, "Learning
recognition and segmentation of 3-D objects from 3D images," in Proceedings of the International
Conference on Computer Vision (ICCV 93). Berlin,
Germany, 1993, pp.121-128.
[154] T. Kohonen, "The self-organizing map," Proceedings
of the IEEE, Vol.78, pp.1464-1480, 1990.
[155] T. Kohonen, Self-organizing maps. Berlin, Germany:
Springer-Verlag, 1995.
[156] A. Eleyan and H. Demirel, "Face Recognition
System Based on PCA and Feedforward Neural Networks," in Computational Intelligence and Bioinspired
Systems, Vol.3512, Lecture Notes in Computer Science:
Springer Berlin / Heidelberg, 2005, pp.935-942.
[157] B. Li and H. Yin, "Face Recognition Using RBF
Neural Networks and Wavelet Transform," in
Advances in Neural Networks – ISNN 2005, vol.3497,
Lecture Notes in Computer Science: Springer Berlin /
Heidelberg, 2005, pp.105-111.
[158] P. Melin, C. Felix, and O. Castillo, "Face recognition
using modular neural networks and the fuzzy Sugeno
integral for response integration," International Journal
Of Intelligent Systems, vol.20, pp.275-291, 2005.
[159] G. C. Zhang, X. S. Huang, S. Z. Li, Y. S. Wang, and
X. H. Wu, "Boosting local binary pattern (LBP)based face recognition," in Advances In Biometric
Person Authentication, Proceedings, Vol.3338, Lecture
Notes In Computer Science, 2004, pp.179-186.
[160] T. Ojala, M. Pietikainen, and M. Maenpaa, "Multiresolution gray-scale and rotation invariant texture
classification width local binary patterns," IEEE
Transactions on Pattern Analysis and Machine
Intelligence, Vol.24, pp.971–987, 2002.
[161] Y. Freund and R. E. Schapire, "A decision-theoretic
Rabia Jafri and Hamid R. Arabnia
[162]
[163]
[164]
[165]
[166]
[167]
[168]
[169]
[170]
[171]
[172]
[173]
generalization of on-line learning and an application
to boosting," Journal of Computer and System
Sciences, Vol.55, pp.119-139, 1997.
T. Ahonen, A. Hadid, and M.Pietikainen, "Face
recognition with local binary patterns," in Proceedings
of the European Conference on Computer Vision,
Vol.3021, Lecture Notes in Computer Science. Prague,
Czech Republic: Springer, 2004, pp.469-481.
U. Krebel, "Pairwise classification and support vector
machines," Advance in Kernel Methods – Support
Vector Learning, pp.255-268, 1999.
T. Hastie and R. Tibshirani, "Classification by
Pairwise Coupling," The Annals of Statistics, Vol.26,
pp.451-471, 1998.
M. Moreira and E. Mayoraz, "Improved Pairwise
Coupling Classification with Correcting Classifiers,"
in Proceedings of the 10th European Conference on
Machine Learning, Vol.1398, Lecture Notes In
Computer Science. London, UK: Springer-Verlag,
1998, pp.160-171.
H. Li, F. Qi, and S. Wang, "Face Recognition with
Improved Pairwise Coupling Support Vector Machines,"
in Computational Intelligence and Bioinspired Systems,
Vol.3512, Lecture Notes in Computer Science: Springer
Berlin / Heidelberg, 2005, pp.927-934.
Z. Li and S. Tang, "Face Recognition Using
Improved Pairwise Coupling Support Vector Machines,"
in Proc. of Intl. Conf. on Neural Information
Processing, Vol.2, 2002, pp.876-880.
J. Platt, "Probabilistic Outputs for Support Vector
Machines and Comparison to Regularized Likelihood
Methods," in Advances in Large Margin Classifiers,
A. J. Smola, P. L. Bartlett, B. Sch¨olkopf, and D.
Schuurmans, Eds.: MIT Press, 2000, pp.61-74.
H. Q. Li, S. Y. Wang, and F. H. Qi, "Automatic face
recognition by support vector machines," in Combinatorial Image Analysis, Proceedings, Vol.3322,
Lecture Notes In Computer Science, 2004, pp.716-725.
C. J. Burges, "A Tutorial on Support Vector
Machines for Pattern Recognition," Data Mining and
Knowledge Discovery, Vol.2, pp.121-267, 1998.
G. Dai and C. Zhou, "Face Recognition Using
Support Vector Machines with the Robust Feature,"
in Proceedings of IEEE Workshop on Robot and
Human Interactive Communication, 2003, pp.49-53.
O. D´eniz, M. Castrill´on, and M. Hern´andez, "Face
Recognition Using Independent Component Analysis
and Support Vector Machines," Pattern Recognition
Letters, Vol.24, pp.2153-2157, 2003.
K. Jonsson, J. Matas, J. Kittler, and Y. P. Li,
"Learning Support Vector Machines for Face Verifi-
[174]
[175]
[176]
[177]
[178]
[179]
[180]
[181]
[182]
[183]
[184]
63
cation and Recognition," in Proceedings of IEEE
International Conference on Automatic and Gesture
Recognition, 2000, pp.208-213.
G. Guo, S. Li, and C. Kapluk, "Face Recognition by
Support Vector Machines," in Proceedings of the
Fourth IEEE International Conference on Automatic
Face and Gesture Recognition. Washington, DC,
USA, 2000, pp.196-201.
Y. Liang, W. Gong, Y. Pan, W. Li, and Z. Hu, "Gabor
Features-Based Classification Using SVM for Face
Recognition," in Advances in Neural Networks –
ISNN 2005, Vol.3497, Lecture Notes in Computer
Science. Chongqing: Springer, 2005, pp.118-123.
L. R. Rabiner, "A tutorial on Hidden Markov Models
and selected applications in speech recognition," in
Readings in Speech Recognition, vol. 77, A. Waibel
and K. Lee, Eds. San Francisco, CA: Morgan
Kaufmann, 1989, pp.257-285.
F. S. Samaria and A. C. Harter, "Parameterisation of
a stochastic model for human face identification," in
Proceedings of the 2nd IEEE Workshop on Applications of Computer Vision. Sarasota, FL, USA, 1994,
pp.138-142.
F. S. Samaria, "Face recognition using Hidden
Markov Models," Trinity College, University of
Cambridge, Cambridge, UK, Ph. D. Thesis 1994.
A. V. Nefian and M. H. Hayes III, "Face Recognition
using an embedded HMM," in IEEE International
Conference Audio Video Biometric based Person
Authentication, 1999, pp.19-24.
S. Kuo and O. Agazzi, "Keyword Spotting in poorly
printed documents using pseudo 2-D Hidden Markov
Models," IEEE Transactions on Pattern Analysis and
Machine Intelligence, Vol.16, pp.842-848, 1994.
C. Liu and H. Wechsler, "Evolutionary Pursuit and
Its Application to Face Recognition," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.22, pp.570-582, 2000.
H.-L. Huang, H.-M. Chen, S.-J. Ho, and S.-Y. Ho,
"Advanced Evolutionary Pursuit for Face Recognition,"
accepted by Journal of VLSI Signal Processing-Systems
for Signal, Image, and Video Technology, 2006.
J. Lu, K. N. Plataniotis, A. N. Venetsanopoulos, and
S. Z. Li, "Ensemble-based Discriminant Learning
with Boosting for Face Recognition," IEEE Transactions on Neural Networks, Vol.17, pp.166-178,
2006.
J. Lu and K. N. Plataniotis, "Boosting face recognition on a large-scale database," in Proceedings of
IEEE International Conference on Image Processing,
Vol.2. Rochester, NY, 2002, pp.109-112.
64
A Survey of Face Recognition Techniques
[185] R. E. Schapire, "The boosting approach to machine
learning: An overview," in MSRI Workshop Nonlinear
Estimation and Classification, 2002, pp.149-172.
[186] F. Roli and J. Kittler, "Multiple Classifier Systems,
Third International Workshop, MCS 2002, Cagliari,
Italy, June 24-26, 2002, Proceedings," in Lecture
Notes in Computer Science, Vol.2364, Lecture Notes
in Computer Science: Springer Verlag, 2002.
[187] X. Lu, Y. Wang, and A. K. Jain, "Combining
Classifiers for Face Recognition," in Proc. IEEE
International Conference on Multimedia & Expo
(ICME 2003). Baltimore, MD, 2003, pp.13-16.
[188] C. M. Bishop, Neural Networks for Pattern Recognition: Oxford University Press, UK, 1995.
[189] G. L. Marcialis and F. Roli, "Fusion of LDA and
PCA for face recognition," in Proceedings of the
Workshop on Machine Vision and Perception, 8th
Workshop of the Italian Association for Artificial
Intelligence (ALLA 02), 2002.
[190] G. L. Marcialis and F. Roli, "Fusion of LDA and
PCA for face verification," in Proceedings of the
Workshop on Biometric Authentication, Vol.2359, LNCS,
M. Tistarelli, J. Bigun, and A. K. Jain, Eds. Copenhagen,
Denmark: Springer-Verlag, 2002, pp.30-37.
[191] G. L. Marcialis and F. Roli, "Fusion of appearancebased face recognition algorithms," Pattern Analysis
and Applications, Vol.7, pp.151-163, 2004.
[192] B. Achermann and H. Bunke, "Combination of
Classifiers on the Decision Level for Face Recognition," Insitut fur Informatik und angewandte
Mathematik, Universitat Bern, Bern, Germany,
Technical Report IAM-96-002 January 1996.
[193] A. S. Tolba and A. N. Abu-Rezq, "Combined
Classifier for Invariant Face Recognition," Pattern
Analysis and Applications, Vol.3, pp.289-302, 2000.
[194] Y. H. Wan, S. M. Ji, Y. Xie, X. Zhang, and P. J. Xie,
"Video program clustering indexing based on face
recognition hybrid model of hidden Markov model
and support vector machine," in Combinatorial Image
Analysis, Proceedings, Vol.3322, Lecture Notes In
Computer Science, 2004, pp.739-749.
[195] K. C. Kwak and W. Pedrycz, "Face recognition: A
study in information fusion using fuzzy integral,"
Pattern Recognition Letters, Vol.26, pp.719-733, 2005.
[196] T. Murofushi and M. Sugeno, "An interpretation of
fuzzy measures and the Choquet integral as an
integral with respect to a fuzzy measure," Fuzzy Sets
System, Vol.29, pp.201-227, 1988.
[197] J. Haddadnia, K. Faez, and M. Ahmadi, "N-Feature
Neural Network Human Face Recognition," Image
and Vision Computing, Vol.22, pp.1071-1082, 2002.
[198] J. Haddadnia, K. Faez, and P. Moallem, "Neural
network based face recognition with moment invariants," in IEEE International Conference on
Image Processing, Vol.1. Thessaloniki, Greece, 2001,
pp.1018-1021.
[199] J. Haddadnia, M. Ahmadi, and K. Faez, "An
Efficient Method for Recognition of Human Face
Recognition Using Higher Order Pseudo Zernike
Moment Invariant," in The 5th IEEE Int. Conf. on
Automatic Face and Gesture Recognition. Washington,
DC, USA, 2002.
[200] C. Beumier and M. Acheroy, "Automatic Face
Recognition," in Proceedings symposium IMAGING.
Eindhoven, The Netherlands, 2000, pp.77-89.
[201] L. Torres, L. Lorente, and J. Vilà, "Face recognition
using self-eigenfaces," in International Symposium
on Image/Video Communications Over Fixed and
Mobile Networks. Rabat, Morocco, 2000, pp.44-47.
[202] R. Chellappa, C. L. Wilson, and S. Sirohey, "Human
and machine recognition of faces: A survey," Proceedings of the IEEE, Vol.83, pp.705-740, 1995.
[203] A. Howell and H. Buxton, "Towards unconstrained
face recognition from image sequences," in Proceedings
of the Second IEEE International Conference on
Automatic Face and Gesture Recognition, 1996,
pp.224-229.
[204] J. Moody and C. Darken, "Learning with localized
receptive fields," in Proceedings of the 1988
Connectionist Models Summer School, D. Touretzky,
G. Hinton, and T. Sejnowski, Eds.: Morgan Kaufmann,
1988, pp.133-143.
[205] J. Moody and C. Darken, "Fast learning in networks
of locally-tuned processing units," Neural Computation,
Vol.1, pp.281-294, 1989.
[206] S. McKenna and S. Gong, "Combined motion and
model-based face tracking," in Proceedings of British
Machine Vision Conference. Edinburgh, UK, 1996,
pp.755-765.
[207] T. E. de Campos, R. S. Feris, and R. M. Cesar Jr., "A
Framework for Face Recognition from Video Sequences Using GWN and Eigenfeature Selection," in
Workshop on Artificial Intelligence and Computer
Vision. Atibaia, Brazil, 2000.
[208] R. S. Feris, T. E. Campos, and R. M. Cesar Jr.,
"Detection and Tracking of Facial Features in Video
Sequences," in Mexican International Conference on
Artificial Intelligence (MICAI 2000), Vol.1793,
Lecture Notes in Artificial Intelligence. Acapulco,
Mexico: Springer-Verlag, 2000, pp.129-137.
[209] V. Kruger and G. Sommer, "Affine real-time face
tracking using a wavelet network," in ICCV’99
Rabia Jafri and Hamid R. Arabnia
[210]
[211]
[212]
[213]
[214]
[215]
[216]
[217]
[218]
[219]
[220]
Workshop: Recognition, Analysis, and Tracking of
Faces and Gestures in Real-Time Systems. Corfu,
Greece, 1999, pp.141-148.
T. E. de Campos, I. Bloch, and R. M. Cesar Jr.,
"Feature Selection Based on Fuzzy Distances Between Clusters: First Results on Simulated Data," in
Proceedings of International Conference on Advances
on Pattern Recognition-ICAPR'2000, Vol.2013, Lecture
Notes In Computer Science. Rio de Janeiro, Brasil:
Springer-Verlag, 2001, pp.186-195.
R. Duda and P. Hart, Pattern Classification and
Scene Analysis. New York, USA: Wiley, 1973.
A. K. Jain, R. P. W. Duin, and J. Mao, "Statistical
Pattern Recognition: A Review," IEEE Transactions
on Pattern Analysis and Machine Intelligence,
Vol.22, pp.4-37, 2000.
Z. Biuk and S. Loncaric, "Face recognition from
multi-pose image sequence," in Proceedings of 2nd
IEEE R8-EURASIP Int'l Symposium on Image and
Signal Processing and Analysis. Pula, Croatia, 2001,
pp.319-324.
V. Krueger and S. Zhou, "Exemplar-based face
recognition from video," in Computer Vision - ECCV
2002: 7th European Conference on Computer Vision,
Copenhagen, Denmark, May 28-31, 2002. Proceedings,
Part IV, Vol.2353, Lecture Notes in Computer
Science: Springer Berlin / Heidelberg, 2002, pp.732.
S. Zhou, V. Krueger, and R. Chellappa, "Face
Recognition from Video: A CONDENSATION
Approach," in Proc. of Fifth 1EEE International
Conference on Automatic Face and Gesture Recognition. Washington D.C., USA, 2002, pp.221-228.
G. L. Marcialis and F. Roli, "Fusion of face
recognition algorithms for video-based surveillance
systems," in Multisensor Surveillance Systems: The
Fusion Perspective, G. L. Foresti, C. S. Regazzoni,
and P. K. Varshney, Eds.: Kluwer, 2003, pp.235-250.
J. Steffens, E. Elagin, and H. Neven, "Person Spotter
– fast and robust system for human detection,
tracking and recognition," in Proceedings, International Conference on Audio- and Video-Based
Person Authentication, 1999, pp.96-101.
presented at Proceedings of the International
Conferences on Audio- and Video-Based Person
Authentication, 1997-2005.
S. Zhou and R. Chellappa, "Beyond a single still
image: Face recognition from multiple still images
and videos," in Face Processing: Advanced Modeling
and Methods: Academic Press, 2005.
M. C. Chiang and T. E. Boult, "Local Blur Estimation and Super-resolution," in Proceedings, IEEE
[221]
[222]
[223]
[224]
[225]
[226]
[227]
[228]
[229]
[230]
[231]
[232]
[233]
65
Conference on Computer Vision and Pattern Recognition, 1997, pp.821-826.
K. Aizawa, T. Komatsu, and T. Saito, "A scheme for
acquiring very high resolution images using multiple
cameras," in Proceedings of the IEEE International
Conference on Acoustics, Speech and Signal Processing, Vol.3, 1992, pp.289-292.
M. Elad and A. Feuer, "Restoration of a single
superresolution image from several blurred, noisy,
and undersampled measured images," IEEE Trans.
Image Processing, Vol.6, pp.1646-1658, 1997.
M. Berthod, H. Shekarforoush, M. Werman, and J.
Zerubia, "Reconstruction of high resolution 3D
visual information using sub-pixel camera displacements," in IEEE Conference on Computer Vision and
Pattern Recognition, 1994, pp.654-657.
M. H. Yang, D. Kriegman, and N. Ahuja, "Detecting
faces in images: a survey," IEEE Transactions on
Pattern Analysis and Machine Intelligence, Vol.24,
pp.34-58, 2002.
L. Torres, "Is there any hope for face recognition?"
in Proc. of the 5th International Workshop on Image
Analysis for Multimedia Interactive Services (WIAMIS
2004). Lisboa, Portugal, 2004.
A. Tibbalds, "Three Dimensional Human Face
Acquisition for Recognition," Trinity College,
University of Cambridge, Cambridge, Ph. D. Thesis
March 1998.
C. Hesher, A. Srivastava, and G. Erlebacher, "A
novel technique for face recognition using range
imaging," in Proceedings of the 7th IEEE International Symposium on Signal Processing and Its
Applications, Vol.2, 2003, pp.201-204.
G. Gordon, "Face Recognition Based on Depth Maps
and Surface Curvature," in SPIE Proceedings:
Geometric Methods in Computer Vision, Vol.1570,
1991, pp.234-- 247.
Cyberware, "Cyberware Inc.: Electronic Documentation."
Scanners, "3D Scanners Ltd. Electronic Documentation, from http://www.3dscanners.com/.HTML."
U. Castellani, M. Bicego, G. Iacono, and V. Murino,
"3D Face Recognition Using Stereoscopic Vision,"
in Advanced Studies in Biometrics, Vol.3161, Lecture
Notes in Computer Science, M. Tistarelli, J. Bigun,
and E. Grosso, Eds.: Springer Berlin / Heidelberg,
2005, pp.126-137.
S. Lee, G. Wolberg, and S. Y. Shin, "Scattered data
interpolation with multilevel B-Splines," IEEE
Transactions on Visualization and Computer Graphics,
Vol.3, pp.228-244, 1997.
G. Pan, Z. Wu, and Y. Pan., "Automatic 3D face
66
[234]
[235]
[236]
[237]
[238]
[239]
[240]
[241]
[242]
[243]
A Survey of Face Recognition Techniques
verification from range data," in International
Conference on Acoustics, Speech, and Signal Processing, Vol.3, 2003, pp.193-196.
C. Xu, Y. Wang, T. Tan, and L. Quan, "Automatic 3D
face recognition combining global geometric
features with local shape variation information," in
International Conference on Automated Face and
Gesture Recognition, 2004, pp.308–313.
Y. Lee, H. Song, U. Yang, H. Shin, and K. Sohn,
"Local feature based 3D face recognition," in Audioand Video-Based Biometric Person Authentication,
Vol.3546, Lecture Notes in Computer Science:
Springer Berlin / Heidelberg, 2005, pp.909-918.
F. R. Al-Osaimi, M. Bennamoun, and A. Mian,
"Integration of local and global geometrical cues for
3D face recognition," Pattern Recognition, Vol.41,
pp.1030-1040, 2008.
J. Y. Cartoux, J. T. LaPreste, and M. Richetin, "Face
authentication or recognition by profile extraction
from range images," in Proceedings of the Workshop
on Interpretation of 3D Scenes, 1989, pp.194-199.
T. Nagamine, T. Uemura, and I. Masuda, "3D facial
image analysis for human identification," in Proceedings
of International Conference on Pattern Recognition
(ICPR' 1992). The Hague, Netherlands, 1992,
pp.324-327.
C. Li and A. Barreto, "Profile-Based 3D Face
Registration and Recognition," in Advances in
Neural Networks – ISNN 2005, Vol.3497, Lecture
Notes in Computer Science: Springer Berlin /
Heidelberg, 2005, pp.478-488.
Y. Wu, G. Pan, and Z. Wu, "Face Authentication
based on Multiple Profiles Extracted from Range
Data," in 4th International Conference on Audioand Video-based Biometric Person Authentication,
Vol. LNCS-2688, 2003, pp.515-522.
B. Gokberk, A. A. Salah, and L. Akarun, "RankBased Decision Fusion for 3D Shape-Based Face
Recognition," in Proceedings of the 5th International
Conference on Audio and Video-based Biometric
Person Authentication, Vol.3546, Lecture Notes in
Computer Science, 2005, pp.1019-1028.
J.-G. Wang, K.-A. Toh, and R. Venkateswarlu,
"Fusion of Appearance and Depth Information for
Face Recognition," in Fifth International Conference
on Audio- and Video-Based Biometric Person
Authentication, Vol.3546, Lecture Notes in Computer
Science: Springer, 2005, pp.919-928.
F. Tsalakanidou, D. Tzocaras, and M. Strintzis, "Use
of depth and colour eigenfaces for face recognition,"
Pattern Recognition Letters, Vol.24, pp.1427–1435,
2003.
[244] K. Chang, K. Bowyer, and P. Flynn, "Face recognition using 2D and 3D facial data," in Multimodal
User Authentication Workshop. Nice, France, 2003,
pp.25-32.
[245] C. Beumier and M. Acheroy, "Face verification from
3D and grey level clues," Pattern Recognition Letters,
Vol.22, pp.1321-1329, 2001.
[246] P. Besl and N. McKay, "A method for registration of
3D shapes," IEEE Transactions on Pattern Analysis
and Machine Intelligence, Vol.14, pp.239–256, 1992.
[247] T. Papatheodorou and D. Reuckert, "Evaluation of
automatic 4D face recognition using surface and
texture registration," in International Conference on
Automated Face and Gesture Recognition, 2004,
pp.321–326.
[248] X. Lu and A. K. Jain, "Integrating range and texture
information for 3D face recognition," in Proceedings
of 7th IEEE Workshop on Applications of Computer
Vision. Breckenridge, CO, 2005, pp.156-163.
[249] Y. Wang, C. Chua, and Y. Ho, "Facial feature
detection and face recognition from 2D and 3D images,"
Pattern Recognition Letters, Vol.23, pp.1191-1202,
2002.
[250] C. Benabdelkader and P. A. Griffin, "Comparing and
combining depth and texture cues for face recognition," Image And Vision Computing, Vol.23,
pp.339-352, 2005.
[251] A. Ruifrok, A. Scheenstra, and R. C. Veltkamp, "A
Survey of 3D Face Recognition Methods," in Audioand Video-based Biometric Person Authentication,
Vol.3546, Lecture Notes in Computer Science:
Springer Berlin / Heidelberg, 2005, pp.891-899.
[252] K. W. Bowyer, K. Chang, and P. J. Flynn, "A survey
of approaches and challenges in 3D and multi-modal
3D+2D face recognition," Computer Vision and
Image Understanding, Vol.101, pp.1-15, 2006.
[253] L. B. Wolff, D. A. Socolinsky, and C. K. Eveland,
"Quantitative measurement of illumination invariance
for face recognition using thermal infrared imagery,"
in CVPR workshop on Computer Vision Beyond
Visual Spectrum. Kauai, HI, USA, 2001.
[254] R. Cutler, "Face recognition using infrared images
and eigenfaces," University of Maryland at College
Park, College Park, MD, USA, Technical report CSC
989, 1996.
[255] A. Gyaourova, G. Bebis, and I. Pavlidis, "Fusion of
Infrared and Visible Images for Face Recognition,"
in Computer Vision - ECCV 2004, Vol.3024, Lecture
Notes in Computer Science: Springer Berlin /
Heidelberg, 2004, pp.456-468.
Rabia Jafri and Hamid R. Arabnia
[256] A. Selinger and D. A. Socolinsky, "AppearanceBased Facial Recognition Using Visible and Thermal
Imagery: A Comparative Study," Equinox Corporation,
Technical Report 02-01 February 2002.
[257] S.-Q. Wu, W. Song, L.-J. Jiang, S. L. Xie, F. Pan, W.Y. Yau, and S. Ranganath, "Infrared Face Recognition
by Using Blood Perfusion Data," in Audio- and
Video-Based Biometric Person Authentication,5th
International Conference, AVBPA 2005, Hilton Rye
Town, NY, USA, July 20-22, 2005, Proceedings,
Vol.3546, Lecture Notes in Computer Science:
Springer, 2005, pp.320-328.
[258] J. Wilder, P. J. Phillips, C. H. Jiang, and S. Wiener,
"Comparison of Visible and Infra-red Imagery for
Face Recognition," in Proceedings, International
Conference on Automatic Face and Gesture Recognition. Killington, VT, USA, 1996, pp.182-187.
[259] D. Socolinsky, L. Wolff, J. Neuheisel, and C. Eveland,
"Illumination invariant face recognition using
thermal infrared imagery," in IEEE Computer Society
International Conference on Computer Vision and
Pattern Recognition, Vol.1. Kauai, HI, USA, 2001,
pp.527-534.
[260] T. Sim, R. Sukthankar, M. Mullin, and S. Baluja,
"Memory-based Face Recognition for Visitor Identification," in Proceedings of the IEEE International
Conference on Automatic Face and Gesture Recognition, 2000, pp.214-220.
[261] X. Chen, P. Flynn, and K. Bowyer, "IR and Visible
Light Face Recognition," Computer Vision and
Image Understanding, Vol.99, pp.332-358, 2005.
[262] X. Chen, P. Flynn, and K. Bowyer, "Visible-light and
infrared face recognition," in Proceedings of the
Workshop on Multimodal User Authentication. Santa
Barbara, California, USA, 2003, pp.48–55.
[263] Identix, "Identix Inc.: Electronic Documentation."
[264] J. Kittler, M. Hatef, R. P. W. Duin, and J. Matas, "On
Combining Classifiers," IEEE Transactions on Pattern
Analysis and Machine Intelligence, Vol.20, pp.226239, 1998.
[265] R. Singh, M. Vatsa, and A. Noore, "Hierarchical
fusion of multi-spectral face images for improved
recognition performance," Information Fusion, Vol.9,
pp.200-210, 2008.
[266] J. Heo, S. Kong, B. Abidi, and M. Abidi, "Fusion of
visual and thermal signatures with eyeglass removal
for robust face recognition," in Proceedings of the
IEEE Workshop on Object Tracking and Classification Beyond the Visible Spectrum in Conjunction
with CVPR, 2004, 2004, pp.94-99.
[267] R. Singh, M. Vatsa, and A. Noore, "Integrated multi-
67
level image fusion and match score fusion of visible
and infrared face images for robust face recognition," Pattern Recognition, Vol.41, pp.880-893, 2008.
[268] S. G. Kong, J. Heo, B. R. Abidi, J. Paik, and M. A.
Abidi, "Recent advances in visual and infrared face
recognition - a review," Computer Vision And Image
Understanding, Vol.97, pp.103-135, 2005.
68
A Survey of Face Recognition Techniques
Rabia Jafri
She received a B.S. degree in Mathematics and Physics
from Islamia University, Pakistan in 1995 followed by B.S.
and Ph.D. degrees from the University of Georgia, U.S.A.
in 1999 and 2008, respectively. Her research interests
include face recognition, gait recognition, multi-biometric
classification systems and image processing.
Hamid R. Arabnia
Hamid R. Arabnia received a Ph.D. degree in Computer
Science from the University of Kent (Canterbury, England)
in 1987. He is currently a Full Professor of Computer
Science at University of Georgia (Georgia, USA), where he
has been since October 1987. His research interests include
Parallel and distributed processing techniques and algorithms,
interconnection networks, and applications. He has chaired
national and international conferences and technical
sessions in these areas; he is the chair of WORLDCOMP
annual research congress. He is Editor-in-Chief of The
Journal of Supercomputing (Springer) and is on the editorial
and advisory boards of 26 other journals and magazines.
He has received a number of awards, including, the
Distinguished Service Award "in recognition and appreciation
of his contributions to the profession of computer science
and his assistance and support to students and scholars
from all over the world" and an "Outstanding Achievement
Award in Recognition of His Leadership and Outstanding
Research Contributions to the Field of Supercomputing".
Prof. Arabnia has published extensively in journals and
refereed conference proceedings. He has over 300 publications
(journals, proceedings, editorship, editorials) in his area of
research. He has been a Co-PI on $7,139,525 externally
funded projects/initiatives (mainly via Yamacraw - includes
some UGA matching). He has also contributed projects for
justification for equipment purchase (grant proposals worth
over $3 Million). In addition, during his tenure as Graduate
Coordinator of Computer Science (August 2002 - January
2009), he secured the largest level of funding in the history
of the Department for supporting the research and
education of graduate students (PhD, MS). Prof. Arabnia
has delivered numerous number of keynote lectures at
international conferences; most recently at (since September
2008): The 14th IEEE International Conference on Parallel
and Distributed Systems (ICPADS'08, Australia); International
Conference on Future Generation Communication and
Networking (FGCN 2008 / IEEE CS, Sanya/China); The
10th IEEE International Conference on High Performance
Computing and Communications (HPCC-08, Dalian/China).
He has also delivered a number of "distinguished lectures"
at various universities.