CURRENT TRENDS ON FACE RECOGNITION …gmail.com, [email protected], [email protected] ABSTRACT: Like all biometrics solutions, face recognition technology measures

International Journal of Computer Engineering and Applications,

Volume XII, Issue I, Jan. 18, www.ijcea.com ISSN 2321-3469

Manjunatha Hiremath, Manukumar N.M, and Venkateswari 153

CURRENT TRENDS ON FACE RECOGNITION METHODS IN

FACE BIOMETRICS

Manjunatha Hiremath1, Manukumar N.M1, Venkateswari2 1 Department of Computer Science, Christ University, Bangalore, 560029

2Research Assistant, Christ University, Bangalore, 560029

[email protected], [email protected], [email protected]

ABSTRACT:

Like all biometrics solutions, face recognition technology measures and matches the unique characteristics for the purposes of identification or authentication. Often leveraging a digital or connected camera, facial recognition software can detect faces in images, quantify their features, and then match them against stored templates in a database. Facial recognition doesn’t just deal with hard identities, but also has the ability to gather demographic data on crowds. This has made face biometrics solutions much sought after in the retail marketing industry. In this paper , we presented a comprehensive review of Face recognition system and the face datasets used in the face biometric.

Keywords: Face Recognition, PCA, ICA, LDA, Kernel, EBGM, EP, Trace transform, AAM, 3D

[1] INTRODUCTION

Facial recognition is a type of biometric procedure that can identify a specific individual in

a digital image by analysing and comparing patterns. So many computer users are notoriously apt

to use poor, easily guessed credentials (such as password), resulting in break-ins where intruders

can guess another user’s credentials and gain unauthorized access to a digital system. The term

biometrics can resolve all type of credential problems by requiring an additional credential

something associated with the person’s own body. The primary intention of this application is to

prevent the frauds from accessing secured resources. This Biometrics offers reliable solution

mailto:[email protected]

CURRENT TRENDS ON FACE RECOGNITION METHODS IN FACE BIOMETRICS


through some technologies. As part of biometrics, the face recognition plays a very important role.

A facial recognition system is a computer application capable of identifying or verifying a person

from a digital image or a video frame from a video source. In this article, we discussed so many

state of the art face recognition procedures with their performance.

[2] FACE RECOGNITION METHODOLOGIES

A. Principal Component Analysis (PCA)

It is one of the statistical technique for reducing the dimensionality of the dataset, and possibly

to feature selection. PCA based algorithm have been the base of the several research projects in

computer vision. Basically PCA is the unsupervised learning technique used for the producing the

optimal linear least square decomposition of a datasets. PCA is a typical de correlation repetition in

present research, one originates an orthogonal projection basis which directly leads to

dimensionality reduction for classification problems.

B. Independent Component Analysis (ICA)

There are a number of algorithms for performing ICA [11], [13], [14], [25]. Author chose the

infomax algorithm proposed by Bell and Sejnowski [11], which was derived from the principle of

optimal information transfer in neurons with sigmoidal transfer functions [27]. The algorithm is

motivated as follows: Let X be an -dimensional (n -D) random vector representing a distribution

of inputs in the environment. (Here, boldface capi-tals denote random variables, whereas plain text

capitals denote matrices). Let W be an n×n invertible matrix, U=WX and Y=f(U) an n-D random

variable representing the outputs of n-neurons. Each component of f = (f1,….,fn) is an invertible

squashing function, mapping real numbers into the interval. Typically, the logistic function is

used

fi (u) = . (1)

The U1.,..,Un variables are linear combinations of inputs and can be interpreted as presynaptic

activations of -neurons. The Y1.,..,Yn variables can be interpreted as postsynaptic activa-tion rates

and are bounded by the interval . The goal in Bell and Sejnowski’s algorithm is to maximize the

mutual informa-tion between the environment X and the output of the neural network Y . This is

achieved by performing gradient ascent on the entropy of the output with respect to the weight

matrix . The gradient update rule for the weight matrix, is as follows:

of the nonlinear where transfer function is the same as the cumulative density

functions of the underlying ICs (up to scaling and translation) it can be shown that maximizing the

joint entropy of the outputs in also minimizes the mutual information be-tween the individual

outputs in [12], [42]. In practice, the logistic transfer function has been found sufficient to separate

mixtures of natural signals with sparse distributions including sound sources [11].

The algorithm is speeded up by including a “sphering” step prior to learning [12]. The row means

of are subtracted, and then is passed through the whitening matrix, which is twice the inverse

square root2 of the covariance matrix




This removes the first and the second-order statistics of the data; both the mean and covariances are

set to zero and the variances are equalized. When the inputs to ICA are the “sphered” data, the full

transform matrix is the product of the sphering ma-trix and the matrix learned by ICA

C. Linear Discriminant Analysis (LDA):

Both PCA and LDA have been used for face recognition [5, 6, 7, 8, 15, 16, 11]. With PCA, the3

input face images usually needed to be w.arped to a standard face because of the large within-class

variance [6, 7]. This processing stage reduces the within-class variance dramatically, thus

improving the recognition rate. Author first built a simple system based on pure LDA [8], but the

performance was not satisfactory on a large dataset of person not present in the training set. The

idea of combining PCA and LDA has been previously explored by Weng et al [15]. Although the

pure LDA algorithm does not have any problem discriminating the trained samples, Author have

observed that it does not perform very well for the following three cases:

• When the testing samples are from persons not in the training set

• When markedly different samples of trained classes are presented

• Samples with different background are presented

• Basically this is a generalization problem since the pure LDA based system is very much tuned to

the specific training set, which has the same number of classes as persons, with 2 or 4 samples per

class!

Combining PCA and LDA, Author obtain a linear projection which maps the input image x first

into the face-subspace y, and then into the classification space z:

y = T X (6) z = WyT y

(7) z = WxT x (8)

where is the PCA transform, Wy is the best linear discriminating transform on PCA feature

space, and Wx is the composite linear projection from the original image space to the classification

space. After this composite linear projection, recognition is performed in the classification space

based on some distance measure criterion.

One thing need to be pointed out is that for the im-plementation of PCA, in many cases, even though

the covariance matrix is a full-rank matrix, the large con-dition number will create a numerical

problem. One way around this is to compute the eigenvalues andeigenvectors for C+I instead of C,

where is a pos-itive number. This is based on the following lemma:Lemma 1 Matrices C and C + I

have same eigen-vectors but different eigenvalues with the relationship: C+I = + as long as + is

not equal to zero.

Performance improvement of this method over pure LDA based method is demonstrated through

our own experiments and FERET test. Author believe that by combining PCA and LDA, using PCA

to construct a task-specific sub-space and then applying LDA on that subspace, other image

recognition systems such as fingerprint, optical character recognition can be improved. Author will

study the subspace-LDA approach in detail and explore the possible applications in future work.



D. Eigenspace-based adaptive approach (EP) :

The task of EP is to search for a face basis through the rotated axes defined in a properly whitened

PCA space. Evolution is driven by a fitness function defined in terms of performance accuracy and

class separation (scatter index). Accuracy indicates the extent to which learning has been successful

so far, while the scatter index gives an indication of the expected fitness on future trials. Together,

the accuracy and the scatter index give an indication of the overall performance ability. In analogy

to the statistical learning theory [43], the scatter index is the conceptual analog for the capacity of

the classifier and its use is to prevent overfitting. By combining these two terms together (with

proper weights), GA can evolve balanced results and yield good recognition performance and

generalization abilities.

One should also point out that just using more principal components (PCs) does not necessarily lead

to better performance, since some PCs might capture the within class scatter which is unwanted for

the purpose of face recognition [25],[27]. In proposed method, one can search the 20- and 30-

dimensional whitened PCA spaces corresponding to the leading eigenvalues, since it is in those

spaces that most of the variations characteristic of human faces occur.

In analogy to pursuit methods, EP seeks to learn an optimal basis for the dual purpose of data

compression and pattern classification. The challenge for EP is to increase the generalization ability

of the learning machine as a result of seeking the trade-off between minimizing the empirical risk

encountered during training and narrowing the confidence interval for reducing the guaranteed risk

for future testing on unseen images. Towards that end, EP implements strategies characteristic of

genetic algorithms (GAs) for searching the space of possible solutions and determining an optimal

basis. Within the face recognition framework, EP seeks an optimal basis for face projections suitable

for compact and efficient face encoding in terms of both present and future recognition ability.

Experimental results, using a large and varied subset from the FERET facial database, show that

the EP method compares favorably against two popular methods for face recognition Eigenfaces

and Fisherfaces.

E. Elastic Bunch Graph Matching (EBGM) :

A first set of graphs is generated manually. Nodes are located at fiducial points and edges between

the nodes as well as correspondences between nodes of different poses are defined. Once the system

has an FBG (possibly consisting of only one manually defined model), graphs for new images can

be generated automatically by elastic bunch graph matching. Initially, when the FBG contains only

few faces, it is necessary to review and correct the resulting matches, but once the FBG is rich

enough (approximately 70 graphs) one can rely on the matching and generate large galleries of

model graphs automatically. Matching a FBG on a new image is done by maximizing a graph

similarity between an image graph and the FBG of identical pose. It depends on the jet similarities

and a topography term, which takes into account the distortion of the image grid relative to the FBG

grid.




Fig. 1. The Face Bunch Graph (FBG) serves as a general representation of faces. Each stack of

discs represents a jet. From a bunch of jets attached to a single node only the best fitting one is

selected for a match, indicated by gray shading.

Fig. 2. Object-adapted grids for different poses. The nodes are positioned automatically by elastic

bunch graph matching.

.

The system presented is general and flexible. It is designed for an in-class recognition task, i.e., for

recognizing members of a known class of objects. Authors have applied it to face recognition but

the system is in no way specialized to faces and author assume that it can be directly applied to

other in-class recognition tasks, such as recognizing individuals of a given animal species, given

the same level of standardization of the images. In contrast to many neural network systems, no

extensive training for new faces or new object classes is required. Only a moderate number of

typical examples have to be inspected to build up a bunch graph, and individuals can then be

recognized after storing a single image.

The system presented is general and flexible. It is designed for an in-class recognition task, i.e., for

recognizing members of a known class of objects. Authors have applied it to face recognition but

the system is in no way specialized to faces and author assume that it can be directly applied to

other in-class recognition tasks, such as recognizing individuals of a given animal species, given

the same level of standardization of the images. In contrast to many neural network systems, no

extensive training for new faces or new object classes is required. Only a moderate number of

typical examples have to be inspected to build up a bunch graph, and individuals can then be

recognized after storing a single image.

Author tested the system with respect to rotation in depth and differences in facial expression. Some

experiments included mirror reflection. Author did not investigate robustness to other variations,

such as illumination changes or structured background. The performance is high on faces of same



pose. Author also showed robustness against rotation in depth up to about 22$. For large rotation

angles the performance degrades significantly. Our system perfoms well compared to other systems.

Results of a blind test of different systems on the FERET database were published in [6] and [7].

The Trace transform, a generalization of the Radon transform, is a new tool for image processing

which can be used for recognizing objects under transformations, e.g. rotation, translation and

scaling. To produce the Trace transform one computes a functional along tracing lines of an image.

Different Trace transforms can be produced from an image using different trace functionals.

The Trace transform [1], a generalization of the Radon transform, is a new tool for image processing

which can be used for recognizing objects under transformations, e.g. rotation, translation and

scaling. To produce the Trace transform one computes a functional along tracing lines of an image.

Each line is characterized by two parameters, namely its distance p

from the centre of the axes and

the orientation the normal to the line has with respect to the reference direction. In addition,

Author define parameter t along the line with its origin at the foot of the normal. The definitions of

these three parameters are shown in figure 1. The image is transformed to another image with the

Trace transform which is a 2-D function depending on parameters ( , p)

. Different Trace

transforms can be produced from an image using different trace functionals. An example of the

Trace transform is shown in figure 2. It is shown that the image space in the x and

y directions is

transformed to the Trace transform space in the and p

directions.

Fig. 3. Tracing line on an image with parameters , p

and t

.

Fig. 4. An image and its Trace transform.




One of the key properties of the Trace transform is that it can be used to construct features invariant

to rotation, translation and scaling. Author should point out that invariance to rotation and scaling

is harder to achieve than invariance to translation. Let us assume that an object is subjected to linear

distortions, i.e. rotation, translation and scaling. It is equivalent to saying that the image remains the

same but viewed from a linearly distorted coordinate system. Consider scanning an image with lines

in all directions. Let us denoted the set of all these lines with . The Trace transform is a function

g defined on with the help of T which is some functional of the image function when it is

considered as a function of variable t . T is called the trace functional.

g( , p) T F( ( , p t, )), (1) where F( , p t, )

stands for the values of the image function

along the chosen line. Parameter t is eliminated after taking the trace functional. The result is

therefore a 2-D function of parameters and p

and can be interpreted as another image defined on

.

F. Active Appearance Model (AAM) :

An Active Appearance Model (AAM) is an integrated statistical model which combines a model of

shape variation with a model of the appearance variations in a shape-normalized frame. An AAM

contains a statistical model if the shape and gray-level appearance of the object of interest which

can generalize to almost any valid example. Matching to an image involves finding model

parameters which minimize the difference between the image and a synthesized model example

projected into the image.

G. 3-D Face Recognition:

Human face is a surface lying in the 3-D space intrinsically. Therefore the 3-D model should be

better for representing faces, especially to handle facial variations, such as pose, illumination etc.

Blantz et al. proposed a method based on a 3-D morphable face model that encodes shape and

texture in terms of model parameters, and algorithm that recovers these parameters from a single

image of a face.

The main novelty of this approach is the ability to compare surfaces independent of natural

deformations resulting from facial expressions. First, the range image and the texture of the face are

acquired. Next, the range image is preprocessed by removing certain parts such as hair, which can

complicate the recognition process. Finally, a canonical form of the facial surface is computed. Such

a representation is insensitive to head orientations and facial expressions, thus significantly

simplifying the recognition procedure. The recognition itself is performed on the canonical surfaces.

H. Bayesian Framework :

A probabilistic similarity measure based on Bayesian belief that the image intensity differences are

characteristic of typical variations in appearance of an individual. Two classes of facial image

variations are defined: intrapersonal variations and extrapersonal variations. Similarity among faces

is measures using Bayesian rule.



All of the face recognition systems cited above (indeed the majority of the face recognition system

published in the open literature) relay on similarity matrics which are invariably based on Euclidean

distance or normalized correlation, thus corresponding to standard “template-matching”-i.e,,.,

nearest-neighbour based recognition. For example, simplest form the similarity measure S I I( 1, 2)

between two facial images I1 and I2

can be set to be inversely proportional to the norm

I. Support Vector Machine (SVM) :

Given a set of points belonging to two classes, a Support Vector Machine (SVM) finds the

hyperplane that separates the largest possible fraction of points of the same class on the same side,

while maximizing the distance from either class to the hyperplane. PCA is first used to extract

features of face images and then discrimination functions between each pair of images are learned

by SVMs.

J. Hidden Markov Models (HMM) :

Hidden Markov Models (HMM) are a set of statistical models used to characterize the statistical

properties of a signal. HMM consists of two interrelated processes: (1) an underlying, unobservable

Markov chain with a finite number of states, a state transition probability matrix and an initial state

probability distribution and (2) a set of probability density functions associated with each state.

K. Boosting & Ensemble Solutions:

The idea behind Boosting is to sequentially employ a weak learner on a weighted version of

a given training sample set to generalize a set of classifiers of its kind. Although any individual

classifier may perform slightly better than random guessing, the formed ensemble can provide a

very accurate (strong) classifier. Viola and Jones build the first real-time face detection system by

using AdaBoost, which is considered a dramatic breakthrough in the face detection research. On

the other hand, papers by Guo et al. are the first approaches on face recogntion using the AdaBoost

methods.

[3] DATA SETS

In biometrics when benchmarking a procedure it is suggested to use a standard test data set for

researchers to be able to directly compare the results. In the state of the art literature there are

numerous datasets in use at present, the choice of an appropriate database to be used should be made

based on the task given (aging, expressions, lighting etc). The other method to choose the data set

specific to the property to be tested (e.g. how algorithm behaves when given images with lighting

deviations or images with diverse facial expressions). Some of the benchmark datasets are given

below:

The Color FERET Database, USA

SCface - Surveillance Cameras Face Database

SCfaceDB Landmarks

Multi-PIE

The Yale Face Database

1 2 I I




The Yale Face Database B

PIE Database, CMU

Project - Face In Action (FIA) Face Video Database, AMP, CMU

AT&T "The Database of Faces" (formerly "The ORL Database of Faces")

Cohn-Kanade AU Coded Facial Expression Database

MIT-CBCL Face Recognition Database

Image Database of Facial Actions and Expressions - Expression Image Database

Face Recognition Data, University of Essex, UK

NIST Mugshot Identification Database

NLPR Face Database

M2VTS Multimodal Face Database (Release 1.00

The Extended M2VTS Database, University of Surrey, UK

The AR Face Database, The Ohio State University, USA

The University of Oulu Physics-Based Face Database

CAS-PEAL Face Database

Japanese Female Facial Expression (JAFFE) Database

BioID Face DB - HumanScan AG, Switzerland

Psychological Image Collection at Stirling (PICS)

The Sheffield Face Database (previously: The UMIST Face Database)

Face Video Database of the Max Planck Institute for Biological Cybernetics

Caltech Faces

EQUINOX HID Face Database

VALID Database

The UCD Colour Face Image Database for Face Detection

Georgia Tech Face Database

Indian Face Database

VidTIMIT Database

Labeled Faces in the Wild

The LFWcrop Database

Labeled Faces in the Wild-a (LFW-a)

3D_RMA database

GavabDB: 3D face database, GAVAB research group, Universidad Rey Juan Carlos,

Spain

FRAV2D Database

FRAV3D Database

BJUT-3D Chinese Face Database

The Bosphorus Database

PUT Face Database

The Basel Face Model (BFM)

Plastic Surgery Face Database

The Iranian Face Database (IFDB)

The Hong Kong Polytechnic University NIR Face Database

The Hong Kong Polytechnic University Hyperspectral Face Database (PolyU-HSFD)

MOBIO - Mobile Biometry Face and Speech Database



Texas 3D Face Recognition Database (Texas 3DFRD)

Natural Visible and Infrared facial Expression database (USTC-NVIE)

FEI Face Database

ChokePoint

UMB database of 3D occluded faces

VADANA: Vims Appearance Dataset for facial ANAlysis

MORPH Database (Craniofacial Longitudinal Morphological Face Database)

Long Distance Heterogeneous Face Database (LDHF-DB)

PhotoFace: Face recognition using photometric stereo

The EURECOM Kinect Face Dataset (EURECOM KFD)

YouTube Faces Database

YMU (YouTube Makeup) Dataset

VMU (Virtual Makeup) Dataset

MIW (Makeup in the "wild") Dataset

3D Mask Attack Database (3DMAD)

Senthilkumar Face Database (Version 1.0)

McGill Real-world Face Video Database

SiblingsDB Database

FaceScrub - A Dataset With Over 100,000 Face Images of 530 People

LFW3D and Adience3D sets

Indian Movie Face database (IMFDB)

Labeled Wikipedia Faces (LWF)

10k US Adult Faces Database

Denver Intensity of Spontaneous Facial Action (DISFA) Database

BU-3DFE Database (Static Data)

BU-4DFE Database (Dynamic Data)

BP4D-Spontanous Database

CAFE - The Child Affective Face Set

UFI - Unconstrained Facial Images

Senthil IRTT Face Database Version 1.1

Senthil IRTT Face Database Version1.2

Senthil IRTT Video Face Database 1.0

VT-AAST Bench-marking Dataset

SEAS-FR-DB (School of Engineering & Applied Science - Face Video Database).

[4] CONCLUSION

Face recognition is a necessity of the modern age as the need for identification of individual has

increased with the globalization of the world. Personal authentication through face has been under

research since last two decades. The performance of the face recognition system has been enhanced

using various algorithms. A generic facial authentication method contains three major steps i.e. face

detection, facial features segmentation and face recognition. There are many commonly used

algorithms used for this purpose. This paper provides an overview of different face recognition

approaches. These approaches are categorized into four classes in this paper. These are holistic

based approach, model based approach, hybrid based approach and feature based approach. Various

techniques introduced in each of these categories are discussed.




REFERENCES

[1] M. Turk, A. Pentland, Eigenfaces for Recognition, Journal of Cognitive Neurosicence, Vol. 3, No. 1, 1991,

pp. 71-86.

[2] M.A. Turk, A.P. Pentland, Face Recognition Using Eigenfaces, Proceedings of the IEEE Conference on

Computer Vision and Pattern Recognition, 3-6 June 1991, Maui, Hawaii, USA, pp. 586-591.

[3] A. Pentland, B. Moghaddam, T. Starner, View-Based and Modular Eigenspaces for Face Recognition,

Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 21-23 June 1994, Seattle,

Washington, USA, pp. 84-91.

[4] H. Moon, P.J. Phillips, Computational and Performance aspects of PCA-based Face Recognition Algorithms,

Perception, Vol. 30, 2001, pp. 303-321.

[5] M.S. Bartlett, J.R. Movellan, T.J. Sejnowski, Face Recognition by Independent Component Analysis, IEEE

Trans. on Neural Networks, Vol. 13, No. 6, November 2002, pp. 1450-1464

[6] K. Etemad, R. Chellappa, Discriminant Analysis for Recognition of Human Face Images, Journal of the

Optical Society of America A, Vol. 14, No. 8, August 1997, pp. 1724-1733

[7] P.N. Belhumeur, J.P. Hespanha, D.J. Kriegman, Eigenfaces vs. Fisherfaces: Recognition using Class Specific

Linear Projection, Proc. of the 4th European Conference on Computer Vision, ECCV'96, 15-18 April 1996,

Cambridge, UK, pp. 45-58

[8] W. Zhao, R. Chellappa, A. Krishnaswamy, Discriminant Analysis of Principal Components for Face

Recognition, Proc. of the 3rd IEEE International Conference on Face and Gesture Recognition, FG'98, 14-16

April 1998, Nara, Japan, pp. 336-341

[9] A.M. Martinez, A.C. Kak, PCA versus LDA, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.

23, No. 2, 2001, pp. 228-233

[10] W. Zhao, A. Krishnaswamy, R. Chellappa, D.L. Swets, J. Weng, Discriminant Analysis of

Principal Components for Face Recognition, Face Recognition: From Theory to Applications, H. Wechsler,

P.J. Phillips, V. Bruce, F.F. Soulie, and T.S. Huang, eds., Springer-Verlag, Berlin, 1998, pp. 73-85

[11] J. Lu, K.N. Plataniotis, A.N. Venetsanopoulos, Face Recognition Using LDA-Based Algorithms, IEEE Trans.

on Neural Networks, Vol. 14, No. 1, January 2003, pp. 195-200

[12] C. Liu, H. Wechsler, Evolutionary Pursuit and Its Application to Face Recognition, IEEE Trans. on Pattern

Analysis and Machine Intelligence, Vol. 22, No. 6, June 2000, pp. 570-582

[13] C. Liu, H. Wechsler, Face Recognition Using Evolutionary Pursuit, Proc. of the Fifth European Conference

on Computer Vision, ECCV'98, Vol II, 02-06 June 1998, Freiburg, Germany, pp. 596-612

[14] L. Wiskott, J.-M. Fellous, N. Krueuger, C. von der Malsburg, Face Recognition by Elastic Bunch Graph

Matching, Chapter 11 in Intelligent Biometric Techniques in Fingerprint and Face Recognition, eds. L.C. Jain

et al., CRC Press, 1999, pp. 355-396

[15] L. Wiskott, J.-M. Fellous, N. Krueuger, C. von der Malsburg, Face Recognition by Elastic Bunch Graph

Matching, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, 1997, pp. 775-779

[16] M.-H. Yang, Kernel Eigenfaces vs. Kernel Fisherfaces: Face Recognition Using Kernel

Methods, Proc. of the Fifth IEEE International Conference on Automatic Face and Gesture

Recognition, 20-21 May 2002, Washington D.C., USA, pp. 215-220

[17] F.R. Bach, M.I. Jordan, Kernel Independent Component Analysis, Journal of Machine Learning Research,

Vol. 3, 2002, pp. 1-48

[18] B. Scholkopf, A. Smola, K.-R. Muller, Nonlinear Component Analysis as a Kernel Eigenvalue Problem,

Technical Report No. 44, December 1996, 18 pages



[19] M.-H. Yang, Face Recognition Using Kernel Methods, Advances in Neural Information Processing Systems,

T. Diederich, S. Becker, Z. Ghahramani, Eds., 2002, vol. 14, 8 pages

[20] S. Zhou, R. Chellappa, B. Moghaddam, Intra-personal kernel space for face recognition, Proc. of the 6th

International Conference on Automatic Face and Gesture Recognition, FGR2004, 17-19 May 2004, Seoul,

Korea, pp. 235-240

[21] S. Zhou, R. Chellappa, Multiple-exemplar discriminant analysis for face recognition, Proc. of the 17th

International Conference on Pattern Recognition, ICPR'04, 23-26 August 2004, Cambridge, UK, pp. 191-194

[22] J. Lu, K.N. Plataniotis, A.N. Venetsanopoulos, Face Recognition Using Kernel Direct Discriminant Analysis

Algorithms, IEEE Trans. on Neural Networks, Vol. 14, No. 1, January 2003, pp. 117-126

[23] A. Kadyrov, M. Petrou, The Trace Transform and Its Applications, IEEE Transactions on Pattern Analysis

and Machine Intelligence, Vol. 23, No. 8, August 2001, pp. 811-828

[24] S. Srisuk, M. Petrou, W. Kurutach and A. Kadyrov, Face Authentication using the Trace Transform,

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

(CVPR'03), 16-22 June 2003, Madison, Wisconsin, USA, pp. 305-312

[25] S. Srisuk and W. Kurutach, Face Recognition using a New Texture Representation of Face

Images, Proceedings of Electrical Engineering Conference, Cha-am, Thailand, 06-07 November 2003, pp.

1097-1102

[26] T.F. Cootes, C.J. Taylor, Statistical Models of Appearance for Computer Vision, Technical Report, University

of Manchester, 125 pages

[27] T.F. Cootes, K. Walker, C.J. Taylor, View-Based Active Appearance Models, Proc. of the IEEE International

Conference on Automatic Face and Gesture Recognition, 26-30 March 2000, Grenoble, France, pp. 227-232

[28] J. Huang, B. Heisele, V. Blanz, Component-based Face Recognition with 3D Morphable Models, Proc. of the

4th International Conference on Audio- and Video-Based Biometric

Person Authentication, AVBPA 2003, 09-11 June 2003, Guildford, UK, pp. 27-34

[29] V. Blanz, T. Vetter, A Morphable Model for the Synthesis of 3D Faces, Proc. of the SIGGRAPH'99, 08-13

August 1999, Los Angeles, USA, pp. 187-194

[30] V. Blanz, T. Vetter, Face Recognition Based on Fitting a 3D Morphable Model, IEEE Transactions on Pattern

Analysis and Machine Intelligence, Vol. 25, No. 9, September 2003, pp. 1063-1074

[31] B. Moghaddam, J.H. Lee, H. Pfister, R. Machiraju, Model-Based 3D Face Capture with Shape-from-

Silhouettes, Proc. of the IEEE International Workshop on Analysis and Modeling of Faces and Gestures,

AMFG, 17 October 2003, Nice, France, pp. 20-27

[32] J. Lee, B. Moghaddam, H. Pfister, R. Machiraju, Finding Optimal Views for 3D Face Shape

Modeling, Proc. of the International Conference on Automatic Face and Gesture Recognition,

FGR2004, 17-19 May 2004, Seoul, Korea, pp. 31-36

[33] A. Bronstein, M. Bronstein, R. Kimmel, and A. Spira. 3D face recognition without facial surface

reconstruction, in Proceedings of ECCV 2004, Prague, Czech Republic, May 11-14, 2004

[34] A. Bronstein, M. Bronstein, and R. Kimmel, Expression-invariant 3D face recognition, Proc. Audio & Video-

based Biometric Person Authentication (AVBPA), Lecture Notes in Comp. Science 2688, Springer, 2003, pp.

62-69

[35] B. Moghaddam, T. Jebara, A. Pentland, Bayesian Face Recognition, Pattern Recognition, Vol. 33, Issue 11,

November 2000, pp. 1771-1782

[36] C. Liu, H. Wechsler, A Unified Bayesian Framework for Face Recognition, Proc. of the 1998

IEEE International Conference on Image Processing, ICIP'98, 4-7 October 1998, Chicago, Illinois, USA, pp.

151-155




[37] B. Moghaddam, C. Nastar, A. Pentland, A Bayesian Similarity Measure for Deformable Image Matching,

Image and Vision Computing, Vol. 19, Issue 5, May 2001, pp. 235-244

[38] G. Guo, S.Z. Li, K. Chan, Face Recognition by Support Vector Machines, Proc. of the IEEE International

Conference on Automatic Face and Gesture Recognition, 26-30 March 2000, Grenoble, France, pp. 196-201

[39] B. Heisele, P. Ho, T. Poggio, Face Recognition with Support Vector Machines: Global versus Component-

based Approach, Proc. of the Eighth IEEE International Conference on Computer Vision, ICCV 2001, Vol.

2, 09-12 July 2001, Vancouver, Canada, pp. 688-694

[40] K. Jonsson, J. Matas, J. Kittler, Y.P. Li, Learning Support Vectors for Face Verification and Recognition,

Proc. of the IEEE International Conference on Automatic Face and Gesture Recognition, 26-30 March 2000,

Grenoble, France, pp. 208-213

[41] A.V. Nefian, M.H. Hayes III, Hidden Markov Models for Face Recognition, Proc. of the IEEE International

Conference on Acoustics, Speech, and Signal Processing, ICASSP'98, Vol. 5, 12-15 May 1998, Seattle,

Washington, USA, pp. 2721-2724

[42] A.V. Nefian, M.H. Hayes, Maximum likelihood training of the embedded HMM for face detection and

recognition, Proc. of the IEEE International Conference on Image Processing, ICIP 2000, Vol. 1, 10-13

September 2000, Vancouver, BC, Canada, pp. 33-36

[43] A.V. Nefian, Embedded Bayesian networks for face recognition, Proc. of the IEEE

International Conference on Multimedia and Expo, Vol. 2, 26-29 August 2002, Lusanne, Switzerland, pp.

133-136

[44] Y. Freund, R.E. Schapire, A Decision-Theoretic Generalization of On-Line Learning and an Application to

Boosting, Journal of Computer and System Sciences, Vol. 55, No. 1, 1997, pp. 119-139

[45] R. Meir, G. Raetsch. An Introduction to Boosting and Leveraging, In S. Mendelson and A. Smola, Editors,

Advanced Lectures on Machine Learning, LNAI 2600, pp. 118-183, Springer, 2003

[46] P. Viola, M.J. Jones, Robust Real-Time Face Detection, International Journal of Computer Vision, Vol. 57,

No. 2, May 2004, pp. 137-154

[47] J. Lu, K.N. Plataniotis, A.N. Venetsanopoulos, S.Z. Li, Ensemble-based Discriminant Learning with Boosting

for Face Recognition, IEEE Transactions on Neural Networks, Vol.

17, No. 1, January 2006, pp. 166-178

[48] G.-D. Guo, H.-J. Zhang, S.Z. Li, Pairwise Face Recognition, Proc. of the Eighth IEEE

International Conference on Computer Vision, ICCV 2001, Vol. 2, 09-12 July 2001, Vancouver, Canada, pp.

282-287

[49] G.-D. Guo, H.-J. Zhang, Boosting for Fast Face Recognition, Second International Workshop on Recognition,

Analysis and Tracking of Faces and Gestures in Real-time Systems, RATFG-

RTS'01, in conjunction with ICCV 2001, 13 July 2001, Vancouver, Canada, pp. 96-100;

Documents

CURRENT TRENDS ON FACE RECOGNITION …gmail.com, [email protected], [email protected] ABSTRACT: Like all biometrics solutions, face recognition technology measures