Face Veriﬁcation using LBP Feature and Clustering

Chenqi Wang, Kevin Lin and Yi-Ping Hung

National Taiwan University, Taipei, Taiwan

Keywords:

Face Recognition, Local Binary Pattern (LBP), Unsupervised Learning.

Abstract:

In this paper, we present a mechanism to extract certain special faces—LBP-Faces, which are designed to

represent different kinds of faces around the world, and utilize them as the basis to verify other faces. In

particular, we show how our idea can integrate with Local Binary Pattern (LBP) and improve its performance.

Other than most of the previous LBP-variant approaches, which, no matter try to improve coding mechanism

or optimize the neighbourhood sizes, ﬁrst divide a face into patch-level regions (e.g. 7× 7 patches), concate-

nating histograms calculated in each patch to derive a rather long dimension vector, and then apply PCA to

implement dimension reduction, our work use original LBP histograms, trying to retain the major properties

such as discriminability and invariance, but in a much bigger component-level region (we divide faces into 7

components). In each component, we cluster LBP descriptors—in the form of histograms to derive N clus-

tering centroids, which we deﬁne as LBP-Faces. Then, to any input face, we calculate its similarities with all

these N LBP-Faces and use the similarities as ﬁnal features to verify the face. It looks like we project the faces

image into a new feature space—LBP-Faces space. The intuition within it is that when we depict an unknown

face, we are prone to use description such as how likely the face’s eye or nose is to an known one. Result of

our experiment on the Labeled Face in Wild (LFW) database shows that our method outperforms LBP in face

veriﬁcation.

1 INTRODUCTION

Face recognition has been an important issue in the

ﬁeld of pattern recognition. This problem has been

addressed on giving two images of face, and veri-

fying that these two images were captured from the

same person or different people, so-called the face

veriﬁcation. This task has been widely applied on the

intelligent surveillance system and become more and

more popular for commercial use. However, most of

the face images captured from the surveillance system

are not ideal. It still has several challenges, such as

the changes of illumination, the variety of head poses,

partially occlusion by addressing the accessories and

so on. Since the face veriﬁcation is a binary classiﬁ-

cation problem, which classiﬁes the given two faces

into same or different people, the most important por-

tion become the feature extraction. Carefully design

a robust and discriminative feature can improve the

performance of face recognition. Thus, the extracted

feature—descriptor is required to be not only discrim-

inativebut also robust to some noise. Among all exist-

ing technologies, local binary pattern(LBP) (Ahonen

et al., 2004) has been demonstrated that it can suc-

cessfully represent the structure of the faces by ex-

Figure 1: Several examples face pairs of the same person

from the Labeled Faces in the Wild data set. Pairs on the

top and bottom are correctly and incorrectly classiﬁed with

our method respectively.

ploiting the distribution of such pixel neighbourhood.

However, those methods suffer from several limita-

tions, such as the ﬁxed quantization and the redundant

feature dimension. It has been argued that a large pro-

portion of the 256 codes in original LBP occur with

a very low frequency, which may cause the code his-

togram less informative and more redundant. Thus, a

lot of LBP varieties have been proposed to improve

the original version.

A famous extension to the original one is known

as Uniform LBP (Ojala et al., 2002), where 256 codes

572

Wang C., Lin K. and Hung Y..

Face Veriﬁcation using LBP Feature and Clustering.

DOI: 10.5220/0004736905720578

In Proceedings of the 9th International Conference on Computer Vision Theory and Applications (VISAPP-2014), pages 572-578

ISBN: 978-989-758-003-1

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

in LBP reduce to 59 codes by merging 198 non-

uniform codes into one bin. The idea stems from the

observation that the codes of non-uniform pattern—

a binary pattern that contains more than two bitwise

transitions from 0 to 1 or vice versa, have very low

frequency to emerge. However, this method is en-

tirely based on the empirical statistics, which seems

too heuristic and has no cogent theory to explain the

reasonableness of mapping each non-uniform code

into a single bin. Additionally, Cao et al. (Cao et al.,

2010) also testify that even in Uniform LBP, some

codes still appear rarely in real-life face images. On

the other hand, when using LBP, most works ﬁrstly

divide a face image into dozens patches and calculate

LBP histograms in each patch before concatenating

all the obtained histograms, which lead to an oversize-

dimension feature vector. To reduce the feature size,

dimension reduction technique such as principle com-

ponent analysis(PCA) should be used to avoid com-

putation load and over-ﬁtting. Despite preserving the

most energy and the largest variation after projection,

PCA may ignore some key discriminative factors.

Instead, our work look at LBP in a totally different

way—from the histogram’s point of view. That is, we

preserve the intact histograms of original LBP and try

to ﬁnd out if there exist some unforeseen yet useful

relations and laws within them. In particular, while

the circular neighbouring pixels can be clustered into

ﬁxed groups in LBP(e.g. if circle size is 8, we can

get 2

codes), we want to know whether histograms

can also be clustered. The intuition is based on the

following observation, that is, when we depict a per-

son, we prefer to use the description such as ‘he has a

hooked nose’ or ‘she has a big eye’. Therefore, con-

sidering LBP histograms are proven to be strong de-

scriptors for face recognition, we divide a image into

7 component, namely eyes, nose, mouth and so on,

and use LBP histogram in each component to repre-

sent its character. Then, we use unsupervised learning

on component basis to get N clustering centroids—

called LBP-Faces, in each component, which we be-

lieve can represent N different kinds of component

around the world if we collect enough data for learn-

ing. Finally, we use the dissimilarities between these

LBP-Faces and each input face image to verify the

people. Figure 1 shows some sample face pairs used

for evaluation in our work. The top pairs are cor-

rectly classiﬁed while the bottom ones not. But we

can see those pairs that we didn’t make right deci-

sion are really difﬁcult to be distinguished, even by

human perceptions. Section 4 illustrates the details of

the dataset—LFW dataset we used in the work. Re-

sult on the dataset shows we have better performance

over LBP. The contribution of this paper includes:

• This paper proposes an innovative idea that in-

spired by the natural recognition procedure of the

human beings to improvethe performance of LBP

in face veriﬁcation.

• To our best knowledge, we are the ﬁrst to use

unsupervised learning in the histogram’s point of

view, which may give a new thought in face veri-

ﬁcation and related recognition ﬁeld.

• Our result on the restricted data set of LFW—a

challenging and authoritativedataset, outperforms

several state-of-the-art face veriﬁcation method,

which prove the rationality and feasibility of our

idea. We believe not only LBP histograms can

be applied on our framework, but some other de-

scriptors may also be applicable.

The rest of this paper is organizedas follows: Sec-

tion 2 discusses the related work in face recognition.

We then describes overview of our proposed method

in Section 3. In Section 4 we introduces the dataset

used in this paper. Section 5 elaborates on how to de-

rive LBP-Faces and how to use these LBP-Faces to

verify people. Our experiment and results are shown

in Section 6 and we conclude our work in Section 7.

2 RELATED WORK

There are a lot of existing approaches to extract fea-

tures in face recognition. Turk et al. (Turk and Pent-

land, 1991) proposed descriptor called EIGENFACE,

where each images is presented as an n-dimension

vector. The input face is projected into the weight

space and the nearest-neighbor method is performed

to ﬁnd the best matched face in the database. FISH-

ERFACE (Belhumeur et al., 1997) is an alternative

method, whose idea is to project the image into a sub-

space in the manner which discounts those regions of

the face with large deviation. Other famous descrip-

tors include discrete cosine transform (DCT) (Rod

et al., 2000) and Gabor (Liu and Wechsler, 2002).

In addition, Ahonen et al. (Ahonen et al., 2004) pro-

posed the Local Binary Pattern (LBP) to represent

features of face and get a reasonable performance. In

detail, a face image is divided into several regions,

and in each region we can derive the LBP histogram.

Finally, the face descriptor is completed by concate-

nating all the histograms into an enhanced feature

vector. However, G. Sharma et al. (Sharma et al.,

2012) pointed out that LBP still has several nonneg-

ligible limitations, such as the baseless heuristic en-

coding mechanisms to reduce the dimension of fea-

ture space, the hard quantization of the feature space,

and the histogram-based feature representation which

FaceVerificationusingLBPFeatureandClustering

573

Figure 2: Pipeline of the proposed unsupervised LBP-Face algorithm.

impeded us to apply higher-order statistics.

Uniform LBP (Ojala et al., 2002) is a common ex-

tension of original one, but Zhimin Cao et al. (Cao

et al., 2010) also argued that even in Uniform LBP,

there are a lot of codes which may rarely appear in

real-life face images. Several methods have been pro-

posed to tackled this problem. Zhimin Cao et al.

introduced the Learning-based (LE) encoder which

projects the ring-based sampled pattern into another

reﬁned feature space. The encoder is trained using

a set of training face images with unsupervised algo-

rithm, such as K-means and random-projection tree.

G. Sharma et al. (Sharma et al., 2012) further pro-

posed the Local Higher-order Statistic (LHS) model,

which improvethe LBP feature by modifying the LBP

feature extraction procedure. They described the lo-

cal pattern of pixel neighbourhoods with their pro-

posed differential vectors, which records the differ-

ence value between the center of the LBP operator

and the circular pixel neighbourhoods. The Gaus-

sian Mixture Model is then applied to describe all of

the differential vectors. Other famous feature-based

methods include local texture pattern(LTP) (Tan and

Triggs, 2010).

Some of the works (Ahonen et al., 2004; Turk and

Pentland, 1991; Belhumeur et al., 1997) mentioned

above focus on how to extract the most discrimina-

tive feature for face recognition. In another aspect,

several previous works (Rod et al., 2000; Cao et al.,

2010) tried to reduce the feature dimension by pro-

jecting the feature into the reﬁned new feature space.

We can easily ﬁnd that there is a trade-off between

the dimension of the feature space and the represen-

tative ability of the feature. The goal of the feature

dimension reduction is to reduce the dimension of the

feature space and decrease the complexity of the com-

putation. However, the more dimension of the feature

we reduce, the more useful information we might lost.

Since there is no theory that can prove which codes

within the histogram are useless for sure, in our work,

we preserve all the histogram codes of LBP to retain

as much representative ability of the initial feature as

possible. Simultaneously, we project the features into

a low-dimension feature space —LBP-Face space red

by using unsupervised learning to fulﬁl dimension re-

duction.

3 OVERVIEW OF FRAMEWORK

In this paper, we propose a novel idea which is in-

spired by the natural recognition procedure of the hu-

man beings, that is, we usually depict a person by

describing how likely he/she is to an known person.

Therefore, we present a mechanism by red ﬁnding out

these ‘well-known’ faces—which we call LBP-Faces,

calculating the dissimilarities between given image

face and all these LBP-Faces, and utilizing the new

distance vectors to verify other unknown faces.

Figure 2 demonstrates the pipeline of our ap-

proach. There are following main steps in our work.

• Firstly, we need to derive LBP-Faces, which is

the most important process in our work since it

results—LBP-Faces would directly inﬂuence our

ﬁnal performance. The upper frame box shows

this process. Speciﬁcally, 13232 face images from

all of world are implemented preprocessing be-

fore extracting their LBP histograms, then we use

unsupervised learning technology to obtain ﬁnal

N centroids, which are denoted as diamond icons

in Figure 2 and can be seen as the most representa-

tive N different faces in the word. The N centroids

are exactly what we call LBP-Faces. Additionally,

in order to preserve local information, all the pro-

cess is carried out in component-level(there are 7

components in each face). For easy description,

we still use LBP-Faces to denote centroids in each

component.

• After acquiring the LBP-Faces, we are ready to

VISAPP2014-InternationalConferenceonComputerVisionTheoryandApplications

574

do face veriﬁcation. Thus, the second step is to

obtain ﬁnal feature for face matching. To any

given face image, we calculate the dissimilarities

between it and all the LBP-Faces to get an N-

dimension distance vector—thisvector is our ﬁnal

feature vector.

• Thus, in face matching stage, we just need to com-

pare these distance vectors among each image.

The smaller the distance of two distance vectors

is, the more chance the two face images are from

the same person. Particularly, since the feature

dimension reduction is simultaneously conducted

within the above process, we don’t need do di-

mension reduction such as PCA like traditional

LBP-variants do, which also makes our method

competitive in computation efﬁciency.

4 DATA SET

Before going further, it is essential to elaborate on

the data set we use in the paper. Since the deriva-

tion of LBP-Faces and our experiment are all based

on the dataset, we use a separate section to illustrate

it—the publicly available dataset Labeled Faces in the

Wild (Huang et al., 2007). Figure 1 in the ﬁrst page

shows some samples of the dataset, which are used in

our experiment in Section 6. This well-known dataset

is speciﬁcally designed for the study of face recogni-

tion. This database contains 13233 face images and

5749 people collected form the web. There are 1680

people have two or more face images for the purpose

of veriﬁcation. All the face images are labeled with

the name of the person. Since the LFW data is col-

lected from the Internet, the images are not always

ideal and become more challenge for face recogni-

tion. That is, there is a large variety in head poses,

different illumination situations, and so on. In order to

eliminate the avoidable variation in the LFW original

version images, the stage of image preprocessing is

performed and is described in detail in subsection 5.1.

In general, the LFW dataset can be separated into

two parts. The ﬁrst part—View 1 is designed for al-

gorithm development, such as model selection or val-

idation. The other part—View 2 is provided for the

performance comparison, which is used for the ﬁnal

evaluation of a proposed algorithm and comparison

of performance of different algorithms. For View 1,

there are 1100 matched pairs and 1100 mismatched

pairs of face images, which can be seen as the positive

and negative samples for the stage of model training.

As the same setting, there are 500 pairs of matched

and mismatched pairs images in View 1, which is

given for the stage of validation. In the conﬁgura-

(a) (b) (c) (d)

Figure 3: (a) input image, (b) face detection result, (c) face

alignment using AAM, and (d) Preprocessing result.

tion of View 2, the database is split into 10 sets ran-

domly. There are 300 matched pairs and 300 mis-

matched pairs images randomly selected within each

set. The ﬁnal evaluation of performance comparison

is given using the 10 folder cross validation, and can

be found in Section 6.

The evaluation results are plotted in the standard

ROC curve. The ROC curve is constructed by plotting

the fraction of True Positive Rate (TPR) and False

Positive Rate (FPR), which illustrates the character-

istic of the binary classiﬁer at various threshold set-

ting. In Section 6, we compare our proposed method

to some state-of-the-art recognition algorithms. The

ROC curves of state-of-the-art are available from

LFW.

5 METHODOLOGY

In this section, we will elaborate on the entire pro-

cess of our work. We ﬁrstly describe some neces-

sary preprocessing before getting LBP-Faces in sub-

section 5.1. Then subsection 5.2 depict in detail how

to derive LBP-Faces. The comparison between two

faces images is described in subsection 5.3.

5.1 Preprocessing

First of all, the haar feature-based cascade classiﬁer

proposed by Viola and Jones (Viola and Jones, 2001)

is applied to detect human faces in the given image,

and extract the region of interest for further process-

ing. Before feature extraction, we need to do some

preprocessing to eliminate avoidable variations and

noise such as illumination, face rotation, and so on.

As to face rotation, since the face image may have

different poses, e.g. the frontal face or left face, we

will perform the component alignment by using Ac-

tive Appearance Model (AAM) (Cootes et al., 2001),

which is proven to be a reliable method in face align-

ment. Then we can resolve the face rotation based on

the line of two detected eyes like Fig 3 shows.

FaceVerificationusingLBPFeatureandClustering

575

0 4 16 64 96 128 160 192 256 512 1024

0.6

0.63

0.66

0.69

0.72

0.75

# of cluster centroids

recognition rate

Eclidean Metric

Cosine Metric

256−dimesion LBP

Figure 4: Performance under different cluster numbers with Euclidean Metric and Cosine Metric.

5.2 Derive LBP-Faces

Our approach is mainly combined with two methods,

LBP and learning-based descriptor. LBP has been

proven to be a both computation efﬁcient and discrim-

inative descriptor. However, it has been argued about

the rationality of its too manual encoding mechanism

and the quite low frequent emergence of a large pro-

portion of codes. We propose to use learning method

to tackle the problem. Unlike most previous works

trying to use learning method in encoding, we think in

a totally different way—we’re learning in histogram’s

point of view. Particularly, we enlarge the region

where a histogram is obtained. The intuition about

this is we think in a larger area, the histogram has

stronger anti-interference ability against the noise. In

other word, the histograms of images from the same

people have more chance to be similar. Therefore, we

can group those similar histograms together and use

a most representative one to stand for this cluster—a

group of similar faces. These representativesare what

we deﬁne LBP-Faces. We use unsupervised learning

techniques to ﬁnd out these LBP-Faces. K-means is

a classic method for exploring clusters by choosing K

centroids to minimize the total distance between them

and their nearest neighbourhoods, whose theory is in

accordance with our idea, so we use it as our default

unsupervised learning algorithm. The centroids are

exactly the LBP-Faces we want to derive.

In unsupervised learning, the choice of the num-

ber of centroids is very important. So we vary the

value of cluster number from 2 to 1024, using default

parameters and initial vectors in K-means, to com-

pare the recognition rate(Face matching will be de-

scribed in next subsection). Also, we compare two

common distance metric, namely Euclidean and Co-

sine distance in unsupervised learning stage. Fig-

ure 4 shows the recognition rate under different clus-

ter numbers using the two distance metric respec-

tively. From the ﬁgure, we can notice that the co-

sine distance metric outperforms the euclidean one,

which corresponds to the point of view proposed by

(Nguyen and Bai, 2011). Thus, we set it as our default

distance metric. It’s worth noting that the accuracy

doesn’t have swift growth as the centroids’ number

increases—when number greater than 16, the curve

ﬂuctuate at a relatively stable level. Therefore, con-

sidering efﬁciency and accuracy, we choose 200 as

our default number of clustering. Moreover, the pur-

ple diamond represents the accuracy of LBP in the

same context, that is, directly extract LBP histograms

in the each component and then concatenate the 7 his-

tograms to form a 1792(7×256)-dimensionvector for

face matching. We use it as a benchmark to evalu-

ate the power of our method. Under the same con-

dition, our initial result achieves higher performance

than LBP, which can prove the reasonableness of our

approach.

5.3 Face Matching using LBP-Faces

After deriving LBP-Faces, we can do face veriﬁca-

tion. Figure 5 shows the process of face matching.

When a pair of face are input, we ﬁrst divide them

into 7 components. It should be necessary to advert

that the histogram icons in the Figure 5 does not rep-

resent LBP histograms but the distances between the

component’s histograms and the N LBP-Faces. So

each histogram icon denotes a N-dimension distance

vector, which can stand for how similar the compo-

nent is to the N LBP-Faces. In face matching, we

simply calculate the distance between the two faces in

component-wise. Certainly, supervised learning can

be utilised here to optimize the weight of Distance 1

to 7 to boost the performance, but for computation

efﬁciency and precise evaluation of our features, we

VISAPP2014-InternationalConferenceonComputerVisionTheoryandApplications

576

Figure 5: Process of face matching.

Table 1: Performance comparison in different parameters of

initial vector ν and distance metric ξ.

Euclidean L1 cosine correlation

sample 0.647 0.653 0.682 0.642

uniform 0.635 0.678 0.662 0.674

cluster 0.658 0.669 0.713 0.685

just sum them and use a optimal threshold in training

data to predict the testing data. Actually, all the ex-

periments in this paper are conducted in this process.

Moreover, we ﬁnd out that optimizing parame-

ters in clustering can also help improve the perfor-

mance. As we all know, the initial vectors in K-means

impact much on the total accuracy, so we compare

the three common initialization mechanisms, namely,

‘sample’, ‘uniform’and ‘cluster’. While ‘sample’ and

‘uniform’ simply select initial vectors at random and

uniformly respectively, ‘cluster’ perform a prelimi-

nary clustering phase on a random 10% subset of the

whole data set to derive the initial vectors. In addition,

as Figure 4 shows that the distance metric also inﬂu-

ence the result, and in K-means algorithm, the dis-

tance metric is extremely important. So we also com-

pare the four common distance metrics—Euclidean

distance, Cosine distance, cityblock(i.e. L1) distance

and correlation distance. All the results are shown in

Table 1, from which, we can ﬁnd that ‘cluster’ method

in initialising vectors and cosine similarity metric in

K-means outperform the other parameters. And the

highest recognition rate is 71.3%. Since the chosen

data set is really challenging, the result is quite satis-

factory.

6 EXPERIMENT AND RESULTS

In this section, we illustrate our experiment and re-

port the ﬁnal face recognition results on LFW bench-

mark. We use View 2 in LFW dataset for evalua-

tion. This data is provided for researchers to evalu-

ate and compare the performance in face recognition.

Although the dataset is ﬁxed, its reasonable construc-

tion by randomly sampling data from a much larger

dataset makes it convinced enough for correct evalua-

tion. To avoid over-ﬁtting by using machine learning

mechanisms with parameters selection and evaluate

the strength and rationality of our proposed features,

we just calculate the dissimilarity of the distance vec-

tors of each pair images as what described in subsec-

tion 5.3. In addition, as subsection 5.3 reveals, we

choose cosine distance measures and ‘cluster’ as de-

fault parameter of distance measures and initial vector

respectively in unsupervised learning to derive the ﬁ-

nal LBP-Faces.

0 0.2 0.4 0.6 0.8 1

0.2

0.4

0.6

0.8

false positive rate

true positive rate

LBP−Face

LBP

LARK−Unsupervised

GJD−BC−100

Eigenface

Figure 6: Face recognition comparison on the restricted

LFW benchmark.

Figure 6 shows the ROC curve and the point on

the curve of our result-the red one, represents the av-

erage over the 10 folds of false positive rate and true

positive rate for a ﬁxed threshold. The other 4 curves

are chosen for comparison. We choose Eigenface be-

cause it has a similar idea with us. We both try to

project a face image into another feature space ac-

cording to some projection basis, but use totally dif-

ferent method. And LBP are classical descriptors and

our method are based on it, thus included. Addi-

tionally, since our model use unsupervised method,

we should choose some unsupervised results to com-

pare. GJD-BC-100 and LARK unsupervised (Ver-

schae et al., 2008) are two unsupervised descrip-

tors enumerated in LFW, which can be benchmarks

to evaluate our work. From the ﬁgure, we can see

our method have much better performance over the

two unsupervised method and Eigenface while out-

perform the LBP a little. Speciﬁcally, we achieve

72.3% accuracy, a little higher than 71.4% of LBP

but much higher than 60.0% of Eigenface. Although

our result has not extremely exciting, it has proventhe

value and rationality of our novel idea.

7 CONCLUSIONS AND FUTURE

WORK

This work constructs the computation efﬁcient face

FaceVerificationusingLBPFeatureandClustering

577

veriﬁcation mechanism. Without any complex ma-

chine learning approach, the designed mechanism

veriﬁes people using only the LBP-Faces and its sim-

ilarity distances. Although the stage of unsupervised

clustering need to take a period of time to process, the

new feature can immediately generated by the LBP-

Faces in the stage of testing. The new feature is cal-

culated only by the similarity distance, which means

the computation complexity is very low and can speed

up the veriﬁcation procedure. Our proposed method

can automatically derive the LBP-Faces and verify-

ing people in nearly real-time, which is applicable in

intelligent mobile phone and embedded system de-

sign. Experimental results show that our method can

achieve higher recognition accuracy than that of the

LBP and Eigenface in the Labeled Faces in the Wild

(LFW). Even though the recognition of our method

might be less promising when the face is partially oc-

cluded or the head pose is severely varied, we believe

that the improvement can be achieved by utilizing the

3D information to enhance the LBP-Faces and clus-

tering the LBP-Faces into different poses. In conclu-

sion, this work is a good initial start, which prove the

reasonableness of our novel idea in face veriﬁcation

and still have a long way to go for future stronger

work.

In this work, we just choose K-means as default

unsupervised learning algorithm, so in the future, we

will ﬁrstly attempt on more clustering mechanisms.

In addition, we will take 3D scenario into consider-

ation in unsupervised learning stage, that is, we will

derive LBP-Faces according to different poses to im-

prove the accuracy. Moreover, the data set—for ex-

ample, the size and the sampling images, used in

learning stage can be also further researched on.

ACKNOWLEDGEMENTS

The work was supported in part by the Image and Vi-

sion Lab at National Taiwan University. The author

was supported by MediaTek Fellowship.

REFERENCES

Ahonen, T., Hadid, A., and Pietikinen, M. (2004). Face

recognition with local binary patterns. In Proc.7th

European Conference on Computer Vision(ECCV),

pages 469–481.

Belhumeur, P., Hespanha, J., and Kriegman, D. (1997).

Eigenfaces vs. ﬁsherfaces: recognition using class

speciﬁc linear projection. IEEE Transactions on Pat-

tern Analysis and Machine Intelligence, 19(7):711–

720.

Cao, Z., Yin, Q., Tang, X., and Sun, J. (2010). Face

recognition with learning-based descriptor. In Proc.

23th IEEE Conference Computer Vision and Pattern

Recognition (CVPR), pages 2707–2714.

Cootes, T., Edwards, G., and Taylor, C. (2001). Active ap-

pearance models. IEEE Transactions on Pattern Anal-

ysis and Machine Intelligence,, 23(6):681–685.

Huang, G. B., Mattar, M., Berg, T., and Learned-miller, E.

(2007). E.: Labeled faces in the wild: A database

for studying face recognition in unconstrained envi-

ronments. Technical report.

Liu, C. and Wechsler, H. (2002). Gabor feature based classi-

ﬁcation using the enhanced ﬁsher linear discriminant

model for face recognition. IEEE Transactions on Im-

age Processing, 11(4):467–476.

Nguyen, H. V. and Bai, L. (2011). Cosine similarity met-

ric learning for face veriﬁcation. In Proc. 10th Asian

conference on Computer vision (ACCV), ACCV’10,

pages 709–720, Berlin, Heidelberg. Springer-Verlag.

Ojala, T., Pietikainen, M., and Maenpaa, T. (2002). Mul-

tiresolution gray-scale and rotation invariant texture

classiﬁcation with local binary patterns. IEEE Trans-

actions on Pattern Analysis and Machine Intelligence,

24(7):971–987.

Rod, Z. P., Adams, R., and Bolouri, H. (2000). Dimension-

ality reduction of face images using discrete cosine

transforms for recognition. In IEEE Conference on

Computer Vision and Pattern Recognition.

Sharma, G., Hussain, S., and Jurie, F. (2012). Local higher-

order statistics (lhs) for texture categorization and

facial analysis. In Proc.15th European Conference

on Computer Vision (ECCV), pages 1–12. Springer

Berlin Heidelberg.

Tan, X. and Triggs, B. (2010). Enhanced local texture fea-

ture sets for face recognition under difﬁcult lighting

conditions. IEEE Transactions on Image Processing,

19(6):1635–1650.

Turk, M. and Pentland, A. (1991). Face recognition using

eigenfaces. In Computer Vision and Pattern Recogni-

tion, 1991. Proceedings CVPR ’91., IEEE Computer

Society Conference on, pages 586–591.

Verschae, R., Ruiz-Del-Solar, J., and Correa, M. (2008).

Face Recognition in Unconstrained Environments: A

Comparative Study. In Workshop on Faces in ’Real-

Life’ Images: Detection, Alignment, and Recognition,

Marseille, France. Erik Learned-Miller and Andras

Ferencz and Fr´ed´eric Jurie.

Viola, P. and Jones, M. (2001). Rapid object detection us-

ing a boosted cascade of simple features. In Proc.

14th IEEE Computer Society Conference on Com-

puter Vision and Pattern Recognition (CVPR), vol-

ume 1, pages I–511–I–518 vol.1.

VISAPP2014-InternationalConferenceonComputerVisionTheoryandApplications

578