I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
6
,
No
.
6
,
Decem
b
er
201
6
,
p
p
.
2
8
0
5
~
2
8
0
9
I
SS
N:
2088
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
6i
6
.
1
3
2
6
4
2805
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
JE
C
E
Face So
ng
P
la
y
er
Accordin
g
t
o
Faci
a
l Ex
press
io
ns
Sa
m
ule Lee
,
Seo
ng
-
Yo
o
n Shi
n
S
c
h
o
o
l
o
f
Co
m
p
u
ter In
f
o
rm
a
ti
o
n
&
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
,
Ku
n
sa
n
Na
ti
o
n
a
l
Un
iv
e
rsit
y
,
Re
p
u
b
li
c
o
f
Ko
re
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
0
7
,
2
0
1
6
R
ev
i
s
ed
No
v
1
0
,
2
0
1
6
A
cc
ep
ted
No
v
24
,
2
0
1
6
Co
n
tem
p
o
ra
ry
p
e
o
p
le
h
a
v
e
h
ig
h
ly
in
su
ff
icie
n
t
ti
m
e
a
n
d
m
e
a
n
s
o
f
re
li
e
v
in
g
th
e
ir
stre
ss
.
P
r
o
v
isio
n
o
f
a
p
ro
g
r
a
m
th
a
t
c
a
n
so
lv
e
su
c
h
stre
ss
i
n
d
a
il
y
li
fe
w
o
u
ld
m
a
k
e
o
n
e
’s
li
f
e
su
b
sta
n
ti
a
ll
y
m
o
re
e
n
jo
y
a
b
le.
In
th
is
th
e
sis,
F
a
c
e
S
o
n
g
P
lay
e
r,
w
h
ich
is
a
sy
ste
m
t
h
a
t
re
c
o
g
n
ize
s
th
e
f
a
c
ial
e
x
p
re
ss
io
n
o
f
a
n
in
d
iv
id
u
a
l
a
n
d
p
lay
s
m
u
sic
th
a
t
is
a
p
p
ro
p
r
iate
f
o
r
su
c
h
p
e
rso
n
,
is
p
re
se
n
ted
.
It
stu
d
ies
i
n
f
o
rm
a
ti
o
n
o
n
th
e
f
a
c
ial
c
o
n
t
o
u
r
li
n
e
s
a
n
d
e
x
trac
ts
a
n
a
v
e
ra
g
e
,
a
n
d
a
c
q
u
ires
th
e
f
a
c
ial
sh
a
p
e
in
f
o
r
m
a
ti
o
n
.
M
UCT
DB
w
a
s
u
se
d
a
s
t
h
e
DB
f
o
r
lea
rn
in
g
.
F
o
r
t
h
e
re
c
o
g
n
it
i
o
n
o
f
f
a
c
ial
e
x
p
re
ss
io
n
,
a
n
a
lg
o
ri
th
m
w
a
s
d
e
sig
n
e
d
b
y
u
sin
g
th
e
d
if
fe
re
n
c
e
s
in
th
e
c
h
a
ra
c
teristics
o
f
e
a
c
h
o
f
th
e
e
x
p
r
e
ss
io
n
s
o
n
th
e
b
a
sis
o
f
e
x
p
re
ss
io
n
les
s
i
m
a
g
e
s.
F
a
c
ial
e
x
p
re
ss
io
n
is
e
x
t
ra
c
ted
b
y
a
c
q
u
iri
n
g
i
n
f
o
r
m
a
ti
o
n
o
n
th
e
e
y
e
s,
e
y
e
b
ro
w
s,
e
y
e
li
d
s,
m
o
u
th
,
li
p
s
a
n
d
n
a
sa
l
c
h
e
e
k
s
f
o
r
e
x
p
re
ss
io
n
s
o
f
h
a
p
p
in
e
ss
,
su
rp
rise
a
n
d
so
rro
w
a
s
w
e
ll
a
s
a
b
se
n
c
e
o
f
e
x
p
re
ss
io
n
.
T
h
e
re
is
a
n
a
d
v
a
n
tag
e
o
f
b
e
in
g
a
b
le
to
o
b
tain
a
su
b
sta
n
ti
a
l
e
ffe
c
t
w
it
h
v
e
r
y
lo
w
c
o
st t
h
ro
u
g
h
th
is
sy
ste
m
.
K
ey
w
o
r
d
:
Face
s
o
n
g
p
la
y
er
Facial
co
n
to
u
r
li
n
es
Facial
ex
p
r
ess
io
n
Facial
s
h
ap
e
in
f
o
r
m
atio
n
MU
C
T
DB
Co
p
y
rig
h
t
©
2
0
1
6
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Seo
n
g
-
Yo
o
n
S
h
i
n
,
Sch
o
o
l o
f
C
o
m
p
u
ter
I
n
f
o
r
m
ati
o
n
&
C
o
m
m
u
n
icat
io
n
E
n
g
i
n
ee
r
in
g
,
Ku
n
s
a
n
Natio
n
al
U
n
i
v
er
s
it
y
,
558
Dae
h
ak
-
r
o
,
Gu
n
s
a
n
-
si
-
5
4
1
5
0
,
R
ep
u
b
lic
o
f
Ko
r
ea
.
E
m
ail: s3
3
9
7
2
2
0
@
k
u
n
s
a
n
.
ac
.
k
r
1.
I
NT
RO
D
UCT
I
O
N
I
t
is
h
i
g
h
l
y
u
n
l
ik
el
y
t
h
at
n
o
o
n
e
a
m
o
n
g
t
h
o
s
e
li
v
i
n
g
i
n
t
h
e
m
o
d
er
n
er
a
ex
p
er
ien
ce
s
n
o
s
tr
ess
.
Stre
s
s
m
an
if
e
s
ts
b
ec
au
s
e
o
f
th
e
n
e
ed
to
a
d
ap
t
to
an
ex
ten
s
i
v
e
r
an
g
e
o
f
ch
a
n
g
e
s
in
a
s
h
o
r
t
p
er
io
d
o
f
tim
e.
L
eis
u
r
eli
n
es
s
is
in
cr
ea
s
in
g
l
y
d
is
ap
p
ea
r
in
g
f
r
o
m
o
u
r
m
o
d
e
r
n
s
o
ciet
y
an
d
p
eo
p
le
ar
e
s
u
b
j
ec
ted
to
m
ar
k
ed
s
tr
ess
e
s
d
u
e
to
n
u
m
er
o
u
s
r
ea
s
o
n
s
in
cl
u
d
in
g
m
o
n
etar
y
is
s
u
e
s
,
s
o
cial
s
tr
u
ctu
r
e
an
d
in
ter
p
er
s
o
n
al
r
elatio
n
s
h
ip
s
,
etc.
I
t
is
th
e
r
ea
lit
y
t
h
at
co
n
te
m
p
o
r
ar
y
p
eo
p
le
h
av
e
lack
o
f
ti
m
e
an
d
m
ea
n
s
o
f
r
elie
v
i
n
g
s
t
r
ess
ar
is
in
g
f
r
o
m
t
h
eir
d
ail
y
liv
e
s
.
A
cc
o
r
d
in
g
l
y
,
w
e
n
ee
d
a
p
r
o
g
r
a
m
t
h
at
ca
n
r
eliev
e
th
e
s
tr
es
s
w
e
en
c
o
u
n
ter
i
n
o
u
r
d
ail
y
lif
e.
Fo
r
ex
a
m
p
le,
w
e
n
ee
d
a
s
y
s
te
m
t
h
at
r
ec
o
g
n
izes
u
s
as
f
r
i
en
d
s
an
d
r
ea
d
s
b
o
o
k
s
b
y
d
is
ce
r
n
in
g
o
u
r
m
o
o
d
.
I
n
ad
d
itio
n
,
it
is
m
u
c
h
m
o
r
e
e
f
f
ec
tiv
e
if
p
s
y
c
h
o
lo
g
ical
p
ac
if
icatio
n
ef
f
ec
t
s
ca
n
b
e
ac
h
i
ev
ed
at
lo
w
co
s
t.
Mo
r
eo
v
er
,
s
u
ch
m
et
h
o
d
w
o
u
ld
b
e
s
u
b
s
ta
n
tiall
y
m
o
r
e
e
f
f
e
ctiv
e
i
f
it
co
u
ld
b
e
u
s
ed
f
o
r
a
d
iv
er
s
e
r
an
g
e
o
f
p
u
r
p
o
s
es in
cl
u
d
in
g
ed
u
ca
tio
n
o
r
g
a
m
es
f
o
r
th
e
u
s
er
s
.
No
t
ev
er
y
b
o
d
y
ex
p
er
ien
ce
s
t
h
e
s
a
m
e
s
tr
es
s
f
r
o
m
th
e
s
a
m
e
ev
en
t.
I
n
f
ac
t,
an
ap
p
r
o
p
r
iate
lev
el
o
f
s
tr
ess
ca
n
b
ec
o
m
e
t
h
e
m
o
ti
v
a
tio
n
to
ac
co
m
p
lis
h
g
o
als,
in
d
u
ce
s
m
o
ti
v
atio
n
a
n
d
p
r
o
v
id
es
v
italit
y
to
li
f
e
b
y
b
ec
o
m
i
n
g
t
h
e
ca
tal
y
s
t
in
li
f
e.
Ho
w
ev
er
,
e
x
ce
s
s
i
v
e
s
tr
ess
d
am
ag
e
s
h
ea
lt
h
b
y
i
m
p
ar
ti
n
g
a
s
u
b
s
ta
n
tial
ad
v
er
s
e
ef
f
ec
t
o
n
t
h
e
b
o
d
y
a
n
d
m
i
n
d
b
ec
au
s
e
h
o
w
o
n
e
d
is
ce
r
n
s
a
n
d
ac
ts
t
o
w
ar
d
s
h
o
w
s
tr
ess
e
f
f
ec
ts
th
e
h
ea
lt
h
an
d
h
ap
p
in
es
s
i
n
o
n
e’
s
l
if
e.
T
h
er
ef
o
r
e,
it
w
o
u
ld
b
e
p
o
s
s
ib
le
f
o
r
p
eo
p
le
to
lead
th
eir
li
v
es
i
n
a
m
o
r
e
b
en
e
f
ic
ial
d
ir
ec
tio
n
if
t
h
e
y
co
u
ld
u
n
d
er
s
t
an
d
th
e
ca
u
s
es o
f
s
tr
es
s
an
d
f
i
n
d
ap
p
r
o
p
r
iate
m
ea
n
s
o
f
r
eliev
in
g
s
tr
es
s
.
Pr
o
d
u
ctio
n
an
d
p
er
ce
p
tio
n
m
u
s
ic
e
n
co
u
r
ag
e
s
a
b
r
o
ad
r
an
g
e
o
f
s
e
n
s
o
r
y
,
co
g
n
iti
v
e
a
n
d
e
m
o
tio
n
al
p
r
o
ce
s
s
es.
E
m
o
tio
n
i
s
th
e
co
r
e
f
u
n
ctio
n
o
f
t
h
e
j
o
y
o
f
m
u
s
ic
.
E
m
o
tio
n
ca
n
b
e
d
ee
m
ed
t
h
e
co
r
e
f
u
n
c
tio
n
o
f
t
h
e
j
o
y
o
f
m
u
s
ic
alo
n
g
w
it
h
t
h
e
m
ar
k
ed
an
d
d
i
v
er
s
i
f
ied
e
m
o
tio
n
al
s
ta
tes
r
ep
o
r
ted
co
n
ti
n
u
o
u
s
l
y
b
y
p
eo
p
le
th
r
o
u
g
h
o
u
t
t
h
e
p
er
io
d
o
f
lis
t
en
in
g
to
m
u
s
ic
[
1
]
.
A
s
s
u
ch
,
m
u
s
ic
i
m
p
ar
t
s
a
s
u
b
s
tan
tial
in
f
l
u
en
ce
o
n
o
u
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
F
a
ce
S
o
n
g
P
la
ye
r
A
cc
o
r
d
in
g
t
o
F
a
cia
l E
x
p
r
ess
io
n
s
(
S
a
mu
le
Lee
)
2806
e
m
o
tio
n
s
.
P
lay
i
n
g
m
u
s
ic
in
ac
co
r
d
an
ce
w
it
h
o
n
e’
s
f
ac
ial
e
x
p
r
ess
io
n
co
u
ld
b
e
v
er
y
in
ter
e
s
tin
g
an
d
b
en
ef
icial
fr
o
m
t
h
e
p
er
s
p
ec
tiv
e
o
f
a
t
h
er
ap
eu
tic
ef
f
ec
t.
S
u
ch
a
s
en
s
ib
le
p
r
o
ce
s
s
is
m
a
n
if
e
s
ted
n
o
t
o
n
l
y
in
m
u
s
ic
b
u
t
also
w
it
h
i
n
a
m
o
n
ito
r
.
E
x
p
a
n
d
in
g
th
e
h
o
r
izo
n
tal
an
g
le
o
f
v
ie
w
in
t
h
e
m
o
n
ito
r
is
a
n
i
m
p
o
r
ta
n
t
ch
ar
ac
ter
i
s
tic
o
f
in
te
g
r
al
i
m
a
g
in
g
m
o
n
ito
r
s
[
2
]
.
T
h
is
m
et
h
o
d
ex
p
a
n
d
s
t
h
e
an
g
le
o
f
v
ie
w
to
b
e
c
h
an
g
ed
b
y
t
h
e
ap
p
r
o
p
r
iate
ap
p
licatio
n
o
f
o
u
r
co
n
v
er
s
io
n
al
g
o
r
ith
m
b
y
i
n
cr
ea
s
i
n
g
t
h
e
d
en
s
it
y
o
f
t
h
e
i
m
a
g
e
ele
m
en
t
i
n
a
h
o
r
izo
n
tal
d
ir
ec
tio
n
.
R
ec
en
t
r
e
s
ea
r
ch
o
n
f
ac
ia
l
r
ec
o
g
n
itio
n
i
n
cl
u
d
e:
[
3
]
th
at
r
ese
ar
ch
ed
th
e
m
u
lti
-
d
ir
e
ctio
n
al
m
u
lti
-
le
v
el
d
u
al
cr
o
s
s
p
atter
n
f
o
r
p
o
w
er
f
u
l
f
ac
ial
r
ec
o
g
n
itio
n
;
[
4
]
th
at
p
r
esen
ted
m
o
b
ile
ap
p
licatio
n
an
d
a
s
y
s
te
m
r
e
f
er
r
ed
to
as
Op
en
Face
,
w
h
ich
i
s
a
w
i
d
ely
u
s
ed
f
ac
ial
r
ec
o
g
n
itio
n
li
b
r
ar
y
;
an
d
[
5
]
th
at
ass
er
ted
th
a
t
th
e
p
u
r
p
o
s
e
o
f
th
e
r
esear
ch
is
f
ac
ial
r
ec
o
g
n
i
tio
n
an
d
e
m
p
h
asized
f
ac
ial
r
ec
o
g
n
i
tio
n
i
n
th
e
co
llectio
n
o
f
t
h
e
tr
ac
k
ed
f
ac
e
s
in
ea
ch
o
f
th
e
p
h
o
to
g
r
ap
h
s
o
r
v
id
eo
.
R
esear
ch
o
n
e
x
p
r
ess
io
n
r
ec
o
g
n
itio
n
also
i
n
cl
u
d
e:
[
6
]
o
n
t
h
e
ch
alle
n
g
e
o
f
r
ec
o
g
n
itio
n
a
n
d
a
n
al
y
s
i
s
o
f
th
e
s
ec
o
n
d
f
ac
ial
ex
p
r
es
s
io
n
w
it
h
t
h
e
s
y
s
te
m
ca
lled
FE
R
A2
0
1
5
;
[
7
]
th
at
p
u
r
s
u
ed
au
to
m
a
tic
f
ac
ial
e
x
p
r
ess
io
n
r
ec
o
g
n
itio
n
b
y
u
s
in
g
a
n
e
x
ce
ll
en
t
f
ac
ial
p
atc
h
f
u
n
ctio
n
;
a
n
d
[
8
]
th
at
p
u
r
s
u
ed
a
s
t
u
d
y
o
n
d
is
cr
i
m
i
n
ati
v
e
s
h
ar
ed
Gau
s
s
ia
n
p
r
o
ce
s
s
es f
o
r
m
u
lt
i
-
v
ie
w
an
d
v
ie
w
-
i
n
v
ar
ia
n
t
f
ac
ial
ex
p
r
ess
io
n
r
ec
o
g
n
itio
n
.
2.
F
ACIAL
RE
CO
G
N
I
T
I
O
N
AND
E
XP
R
E
S
SI
O
N
R
E
CO
G
NI
T
I
O
N
T
h
e
f
ac
e
i
s
r
ec
o
g
n
ized
b
y
u
s
in
g
a
n
A
cti
v
e
Sh
ap
e
Mo
d
el
[
9
]
.
Facial
r
ec
o
g
n
itio
n
,
as
ill
u
s
tr
ated
i
n
Fig
u
r
e
1
,
is
ex
ec
u
ted
b
y
co
m
p
u
ti
n
g
t
h
e
av
er
a
g
e
o
f
t
h
e
in
f
o
r
m
atio
n
o
n
t
h
e
f
ac
ia
l
co
n
to
u
r
li
n
es
t
h
r
o
u
g
h
lear
n
in
g
.
Fig
u
r
e
1
.
Face
R
ec
o
g
n
itio
n
b
y
Sh
ap
e
Fig
u
r
e
2
.
MU
C
T
DB
MU
C
T
DB
[
1
0
]
illu
s
tr
ated
in
Fig
u
r
e
2
w
a
s
u
s
ed
as t
h
e
lear
n
in
g
d
ata
i
n
th
is
t
h
esi
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
2
8
0
5
–
2
8
0
9
2807
E
x
p
r
ess
io
n
r
ec
o
g
n
itio
n
i
s
ex
ec
u
ted
in
th
e
f
o
llo
w
i
n
g
o
r
d
er
.
I
n
th
e
f
ir
s
t
s
e
v
er
al
s
ec
o
n
d
s
,
th
e
d
ata
at
th
e
ti
m
e
o
f
ab
s
en
ce
o
f
ex
p
r
es
s
io
n
is
r
ec
ei
v
ed
,
w
h
ic
h
i
s
t
h
en
u
s
ed
as
t
h
e
S
tan
d
ar
d
to
co
m
p
u
te
t
h
e
d
is
tan
ce
b
et
w
ee
n
ea
ch
o
f
t
h
e
c
h
ar
ac
ter
is
tic
p
o
in
ts
.
Fi
g
u
r
e
3
is
th
e
d
ata
u
s
ed
in
e
x
p
r
ess
io
n
r
ec
o
g
n
it
io
n
.
Fig
u
r
e
3
.
Data
o
f
E
x
p
r
es
s
io
n
R
ec
o
g
n
itio
n
(
1
)
Hap
p
in
ess
:
Di
s
tan
ce
b
et
wee
n
p
h
il
tr
u
m
a
n
d
to
p
o
f
t
h
e
ce
n
ter
o
f
t
h
e
l
ip
s
b
ec
o
m
es
cl
o
s
er
.
T
h
e
d
is
tan
ce
b
et
w
ee
n
b
o
th
th
e
d
is
tal
tip
o
f
t
h
e
m
o
u
th
i
n
cr
ea
s
e
s
.
Heig
h
t o
f
t
h
e
e
y
e
s
d
ec
r
ea
s
es.
T
h
e
b
o
th
d
is
tal
tip
s
o
f
th
e
c
h
e
ek
s
o
f
t
h
e
n
o
s
e
b
ec
o
m
e
b
r
o
ad
er
.
C
o
n
d
itio
n
)
Sta
n
d
ar
d
h
ei
g
h
t o
f
th
e
e
y
e
s
>C
u
r
r
en
t
h
ei
g
h
t o
f
th
e
ey
e
s
A
N
D
Stan
d
ar
d
w
id
t
h
o
f
t
h
e
m
o
u
t
h
<
C
u
r
r
en
t
w
id
th
o
f
th
e
m
o
u
t
h
(
Mo
r
e
th
an
5
%)
A
N
D
Stan
d
ar
d
d
is
tan
ce
b
et
w
ee
n
th
e
p
h
iltru
m
an
d
lip
s
>
C
u
r
r
en
t d
is
tan
ce
b
et
w
ee
n
t
h
e
p
h
il
tr
u
m
an
d
lip
s
(
Mo
r
e
th
an
3
%)
A
N
D
Stan
d
ar
d
d
is
tan
ce
b
et
w
ee
n
th
e
b
o
th
ch
ee
k
s
o
f
th
e
n
o
s
e<
C
u
r
r
en
t
d
is
t
an
ce
b
et
w
ee
n
b
o
th
ch
e
ek
s
o
f
th
e
n
o
s
e
(
Mo
r
e
th
an
3
%)
(
2
)
So
r
r
o
w
:
W
id
th
o
f
th
e
m
id
d
le
o
f
th
e
f
o
r
eh
ea
d
b
ec
o
m
es
n
ar
r
o
w
er
,
e
y
e
s
g
et
s
m
a
l
ler
an
d
th
e
d
is
tal
tip
o
f
t
h
e
m
o
u
t
h
s
a
g
s
.
C
o
n
d
itio
n
s
)
Sta
n
d
ar
d
w
id
t
h
o
f
th
e
m
id
d
le
o
f
th
e
f
o
r
eh
ea
d
>W
id
th
o
f
t
h
e
m
id
d
le
o
f
t
h
e
f
o
r
eh
ea
d
(
Mo
r
e
th
an
1
%)
A
N
D
Stan
d
ar
d
h
ei
g
h
t o
f
th
e
e
y
es>
C
u
r
r
en
t h
e
ig
h
t o
f
t
h
e
e
y
es
A
N
D
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
F
a
ce
S
o
n
g
P
la
ye
r
A
cc
o
r
d
in
g
t
o
F
a
cia
l E
x
p
r
ess
io
n
s
(
S
a
mu
le
Lee
)
2808
Stan
d
ar
d
d
is
tan
ce
b
et
w
ee
n
t
h
e
ch
i
n
an
d
lo
w
er
tip
o
f
th
e
l
ip
s
>C
u
r
r
en
t
b
et
w
ee
n
t
h
e
p
h
il
tr
u
m
an
d
e
y
eb
r
o
w
s
(
Mo
r
e
th
an
3
%)
(
3
)
Su
r
p
r
is
e:
E
y
eb
r
o
w
s
ar
e
r
ai
s
ed
w
it
h
th
e
o
p
en
i
n
g
u
p
o
f
th
e
m
o
u
t
h
.
T
h
e
d
is
tan
ce
b
et
w
ee
n
th
e
e
y
eb
r
o
w
s
an
d
th
e
to
p
o
f
th
e
e
y
elid
s
i
n
cr
ea
s
e
s
.
C
o
n
d
iti
o
n
s
)
Sta
n
d
ar
d
d
is
tan
c
e
b
et
w
ee
n
th
e
p
h
iltru
m
a
n
d
e
y
eb
r
o
w
s
<
C
u
r
r
en
t
d
is
ta
n
ce
b
et
w
ee
n
t
h
e
p
h
iltru
m
a
n
d
e
y
eb
r
o
w
s
(
Mo
r
e
th
an
3
%)
A
N
D
Stan
d
ar
d
h
ei
g
h
t o
f
th
e
m
o
u
t
h
>
C
u
r
r
en
t h
ei
g
h
t o
f
t
h
e
m
o
u
th
A
N
D
Stan
d
ar
d
d
is
tan
ce
b
et
w
ee
n
t
h
e
e
y
eb
r
o
w
s
a
n
d
t
h
e
to
p
o
f
t
h
e
e
y
elid
s
<
C
u
r
r
e
n
t
d
is
ta
n
ce
b
e
t
w
ee
n
t
h
e
e
y
eb
r
o
w
s
a
n
d
th
e
to
p
o
f
th
e
e
y
elid
s
(
Mo
r
e
th
an
3
%)
3.
E
XP
E
R
I
M
E
NT
T
h
e
ex
p
er
im
e
n
t
w
a
s
r
ea
lized
b
y
u
s
i
n
g
Vis
u
al
C
++
2
0
1
3
,
an
d
A
SM
r
ec
o
g
n
itio
n
a
n
d
MU
C
T
D
B
w
er
e
u
s
ed
to
g
et
h
er
.
T
ab
le
1
b
elo
w
i
llu
s
tr
ates t
h
e
A
SM
m
atch
in
g
r
ec
o
g
n
itio
n
r
ates.
T
ab
le
1
.
A
SM
R
ec
o
g
n
itio
n
R
a
tio
N
o
.
o
f
f
a
c
e
s
N
o
.
o
f
A
S
M
r
e
c
o
g
n
i
t
i
o
n
N
o
.
o
f
e
r
r
o
r
s
R
e
c
o
g
n
i
t
i
o
n
r
a
t
e
40
34
6
8
5
%
ASM
m
atc
h
i
n
g
er
r
o
r
s
o
cc
u
r
d
u
e
to
th
e
f
o
llo
w
in
g
r
ea
s
o
n
s
:
a.
W
ea
r
in
g
e
y
e
g
las
s
es
w
i
th
a
t
h
i
ck
f
r
a
m
e
ca
n
b
e
r
ec
o
g
n
ized
as e
y
e
s
.
b.
No
t p
o
s
s
ib
le
to
r
ec
o
g
n
ize
th
e
f
ac
e
if
th
e
h
ea
d
is
s
e
v
er
el
y
t
ilte
d
o
r
tu
r
n
ed
a
w
a
y
.
c.
P
r
o
b
lem
o
f
r
ec
o
g
n
iz
in
g
cr
ea
s
e
s
in
g
ar
m
e
n
ts
,
etc.
as t
h
e
f
ac
e.
d.
R
ec
o
g
n
itio
n
o
f
an
er
r
o
r
if
a
d
if
f
er
en
ce
i
n
li
g
h
t a
n
d
s
h
ad
e
o
cc
u
r
s
d
u
e
to
lig
h
ti
n
g
.
e.
R
ec
o
g
n
itio
n
o
f
o
th
e
r
o
b
j
ec
ts
if
th
e
b
ac
k
g
r
o
u
n
d
is
co
m
p
le
x
.
T
ab
le
2
b
elo
w
i
llu
s
tr
ates
th
e
r
ec
o
g
n
itio
n
r
ate
o
f
ea
c
h
o
f
t
h
e
ex
p
r
ess
io
n
s
.
Alt
h
o
u
g
h
t
h
e
r
ec
o
g
n
itio
n
r
ates
f
o
r
h
ap
p
in
e
s
s
,
s
u
r
p
r
is
e
a
n
d
ex
p
r
es
s
io
n
le
s
s
w
er
e
h
ig
h
,
th
at
f
o
r
a
s
o
r
r
o
w
f
u
l
ex
p
r
es
s
io
n
w
a
s
f
o
u
n
d
to
b
e
lo
w
.
T
h
i
s
i
s
a
p
r
o
b
lem
t
h
at
o
cc
u
r
s
d
u
e
to
t
h
e
lack
o
f
d
is
tin
c
tn
e
s
s
o
f
th
e
ch
ar
ac
ter
i
s
ti
cs
o
f
a
s
o
r
r
o
w
f
u
l
ex
p
r
ess
io
n
i
n
co
m
p
ar
is
o
n
to
o
th
er
ex
p
r
ess
io
n
s
.
I
n
ad
d
itio
n
,
th
er
e
is
t
h
e
is
s
u
e
t
h
at
t
h
e
e
x
p
r
ess
io
n
o
f
s
o
r
r
o
w
d
if
f
er
s
b
et
w
ee
n
i
n
d
iv
id
u
als.
T
ab
le
2
.
E
x
p
r
ess
io
n
R
ec
o
g
n
iti
o
n
R
atio
C
a
t
e
g
o
r
i
e
s
Ex
p
r
e
ssi
o
n
l
e
ss(
1
0
)
H
a
p
p
i
n
e
ss(
1
0
)
S
u
r
p
r
i
se
(
1
0
)
S
o
r
r
o
w
(
1
0
)
N
o
.
o
f
r
e
c
o
g
n
i
t
i
o
n
s
10
8
9
6
N
o
.
o
f
e
r
r
o
r
s
9
2
1
4
R
e
c
o
g
n
i
t
i
o
n
r
a
t
e
9
0
%
8
0
%
9
0
%
6
0
%
4.
CO
NCLU
SI
O
N
I
n
th
i
s
t
h
esi
s
,
Face
So
n
g
P
la
y
er
th
at
r
ec
o
g
n
izes
t
h
e
f
ac
ial
e
x
p
r
ess
io
n
o
f
p
eo
p
le
an
d
p
la
y
i
n
g
m
u
s
ic
th
at
i
s
ap
p
r
o
p
r
iate
f
o
r
th
e
a
m
b
ien
ce
o
f
s
u
c
h
ex
p
r
es
s
io
n
w
a
s
p
r
esen
ted
.
A
n
a
v
er
a
g
e
w
as
ex
tr
ac
ted
th
r
o
u
g
h
lear
n
in
g
i
n
f
o
r
m
atio
n
o
n
f
ac
ial
co
n
to
u
r
lin
es
a
n
d
th
e
ac
q
u
ir
e
d
in
f
o
r
m
atio
n
o
n
f
ac
ia
l
s
h
ap
e
s
.
MU
C
T
DB
w
as
u
s
ed
as
th
e
DB
f
o
r
th
is
p
u
r
p
o
s
e.
Fo
r
th
e
r
ec
o
g
n
itio
n
o
f
th
e
f
ac
ial
ex
p
r
es
s
io
n
s
,
an
alg
o
r
it
h
m
w
as
d
esi
g
n
ed
b
y
u
s
i
n
g
th
e
d
if
f
er
en
ce
s
in
t
h
e
s
p
ec
ial
f
ea
tu
r
es
o
f
t
h
e
ex
p
r
ess
io
n
s
.
A
cc
o
r
d
in
g
l
y
,
f
ac
ial
ex
p
r
ess
io
n
s
w
er
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
6
,
No
.
6
,
Dec
em
b
er
2
0
1
6
:
2
8
0
5
–
2
8
0
9
2809
ex
tr
ac
ted
b
y
ac
q
u
ir
in
g
in
f
o
r
m
atio
n
o
n
th
e
e
y
es,
e
y
el
id
s
,
e
y
eb
r
o
w
s
,
m
o
u
th
,
lip
s
a
n
d
ch
ee
k
s
o
f
t
h
e
n
o
s
e
f
o
r
h
ap
p
y
,
s
u
r
p
r
is
ed
,
s
o
r
r
o
w
f
u
l a
n
d
ex
p
r
ess
io
n
le
s
s
f
ac
es.
RE
F
E
R
E
NC
E
S
[1
]
P
a
tri
k
V
u
il
leu
m
ier
a
n
d
W
ieb
k
e
T
ro
st,
"
M
u
sic
a
n
d
e
m
o
ti
o
n
s:
f
ro
m
e
n
c
h
a
n
tm
e
n
t
to
e
n
train
m
e
n
t"
,
An
n
.
N
.
Y
.
Aca
d
.
S
c
i
.
V
o
l
.
1
3
3
,
p
p
.
2
1
2
–
2
2
2
,
2
0
1
5
.
[2
]
A
d
rián
Do
ra
d
o
,
G
e
n
a
ro
S
a
a
v
e
d
r
a
,
Jo
rg
e
S
o
la
-
P
ik
a
b
e
a
,
a
n
d
M
a
n
u
e
l
M
a
rtí
n
e
z
-
Co
rra
l,
"
In
teg
ra
l
Im
a
g
in
g
M
o
n
it
o
rs
w
it
h
a
n
En
larg
e
d
V
iew
in
g
A
n
g
le
"
,
J
.
ln
f.
Co
mm
u
n
.
Co
n
v
e
rg
.
E
n
g
.
V
o
l
.
1
3
,
N
o
.
2
,
p
p
.
1
3
2
-
1
3
8
,
J
u
n
.
2
0
1
5
.
[3
]
Din
g
,
C.
,
Ch
o
i,
J.
,
T
a
o
,
D.,
&
Da
v
is,
L
.
S
.
,
“
M
u
lt
i
-
d
irec
ti
o
n
a
l
m
u
lt
i
-
lev
e
l
d
u
a
l
-
c
ro
ss
p
a
tt
e
rn
s
f
o
r
ro
b
u
st
f
a
c
e
re
c
o
g
n
it
io
n
”
,
IEE
E
tra
n
s
a
c
ti
o
n
s o
n
p
a
tt
e
rn
a
n
a
lys
is
a
n
d
m
a
c
h
i
n
e
in
telli
g
e
n
c
e
,
3
8
(3
),
5
1
8
-
5
3
1
,
2
0
1
6
.
[4
]
Am
o
s,
B.
,
L
u
d
w
ic
z
u
k
,
B.
,
&
S
a
ty
a
n
a
ra
y
a
n
a
n
,
M
.
,
“
Op
e
n
F
a
c
e
:
A
g
e
n
e
ra
l
-
p
u
rp
o
se
f
a
c
e
re
c
o
g
n
it
io
n
l
i
b
ra
ry
w
it
h
m
o
b
il
e
a
p
p
li
c
a
ti
o
n
s
”
,
2
0
1
6
.
[5
]
P
a
rk
h
i,
Om
k
a
r
M
.
,
A
n
d
re
a
V
e
d
a
ld
i,
a
n
d
A
n
d
re
w
Zi
ss
e
r
m
a
n
.
"
De
e
p
f
a
c
e
re
c
o
g
n
it
io
n
"
,
Brit
ish
M
a
c
h
in
e
Vi
si
o
n
Co
n
fer
e
n
c
e
.
V
o
l
.
1
.
No
.
3
.
2
0
1
5
.
[6
]
V
a
l
sta
r,
M
.
F
.
,
A
l
m
a
e
v
,
T
.
,
G
irar
d
,
J.M
.
,
M
c
Ke
o
w
n
,
G
.
,
M
e
h
u
,
M
.
,
Yin
,
L
.
&
Co
h
n
,
J.
F.
,
In
A
u
to
m
a
ti
c
F
a
c
e
a
n
d
G
e
stu
re
Re
c
o
g
n
it
io
n
(F
G
)
”
,
Fer
a
2
0
1
5
-
se
c
o
n
d
fa
c
i
a
l
e
x
p
re
ss
io
n
re
c
o
g
n
it
i
o
n
a
n
d
a
n
a
lys
is
c
h
a
ll
e
n
g
e
,
2
0
1
5
1
1
th
IEE
E
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
a
n
d
W
o
rk
sh
o
p
s
o
n
,
Vo
l.
6
,
p
p
.
1
-
8
,
2
0
1
5
,
M
a
y
.
[7
]
Ha
p
p
y
,
S
.
L
.
,
&
Ro
u
tray
,
A
.
,
“
Au
to
m
a
ti
c
fa
c
ial
e
x
p
re
s
sio
n
re
c
o
g
n
it
io
n
u
sin
g
f
e
a
tu
re
s
o
f
sa
li
e
n
t
fa
c
ial
p
a
tch
e
s
”
,
IEE
E
tra
n
sa
c
ti
o
n
s
o
n
Af
fec
ti
v
e
Co
mp
u
t
in
g
,
6
(1
),
1
-
12
,
2
0
1
5
.
[8
]
El
e
f
th
e
riad
is,
S
tef
a
n
o
s,
Og
n
jen
R
u
d
o
v
ic,
a
n
d
M
a
ja
P
a
n
ti
c
.
"
Disc
ri
m
in
a
ti
v
e
sh
a
re
d
g
a
u
ss
ian
p
ro
c
e
ss
e
s
f
o
r
m
u
lt
iv
ie
w
a
n
d
v
iew
-
in
v
a
rian
t
f
a
c
ial
e
x
p
re
ss
i
o
n
re
c
o
g
n
it
io
n
"
,
IEE
E
tra
n
s
a
c
ti
o
n
s o
n
ima
g
e
p
r
o
c
e
ss
in
g
2
4
.
1
,
p
p
.
189
-
2
0
4
,
2
0
1
5
.
[9
]
T
i
m
Co
o
tes
,
"
A
n
In
tr
o
d
u
c
ti
o
n
t
o
A
c
ti
v
e
S
h
a
p
e
M
o
d
e
ls
"
,
Ch
a
p
ter
7
:
”
M
o
d
e
l
-
Ba
se
d
M
e
th
o
d
s
i
n
A
n
a
l
y
sis
o
f
Bio
m
e
d
ica
l
I
m
a
g
e
s”
,
in
Ima
g
e
Pro
c
e
ss
in
g
a
n
d
An
a
lys
is
,
Ed
.
R.
Ba
ld
o
c
k
a
n
d
J.G
ra
h
a
m
,
O
x
f
o
rd
Un
iv
e
rsit
y
P
re
ss
,
p
p
2
2
3
-
2
4
8
,
2
0
0
0
[1
0
]
M
il
b
o
rro
w
,
S
.
,
M
o
rk
e
l,
J.
,
Nic
o
ll
s,
F
.
“
T
h
e
M
UCT
L
a
n
d
m
a
rk
e
d
F
a
c
e
Da
tab
a
se
”
,
Pa
tt
e
rn
Re
c
o
g
n
i
ti
o
n
Ass
o
c
ia
ti
o
n
o
f
S
o
u
th
A
frica
,
2
0
1
0
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
S
a
m
u
l
e
Le
e
.
H
e
is
c
u
rre
n
tl
y
a
t
ten
d
i
n
g
f
o
u
rth
g
ra
d
e
in
S
c
h
o
o
l
o
f
Co
m
p
u
ter
a
n
d
In
f
o
rm
a
ti
o
n
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
,
Ku
n
sa
n
Na
ti
o
n
a
l
Un
iv
e
rsity
His In
tere
ste
d
A
re
a
:
a
rti
f
i
c
ial
in
t
e
ll
ig
e
n
c
e
,
m
a
c
h
in
e
lea
rn
in
g
,
c
o
m
p
u
ter v
isio
n
S
e
o
n
g
-
Yo
o
n
S
h
in
.
re
c
e
iv
e
d
h
is
M
.
S
.
a
n
d
P
h
.
D
d
e
g
re
e
s
f
ro
m
th
e
De
p
t.
o
f
Co
m
p
u
ter
In
f
o
rm
a
ti
o
n
En
g
in
e
e
rin
g
o
f
Ku
n
sa
n
Na
ti
o
n
a
l
Un
iv
e
rsit
y
,
Ku
n
sa
n
,
Ko
re
a
,
in
1
9
9
7
a
n
d
2
0
0
3
,
re
sp
e
c
ti
v
e
l
y
.
F
ro
m
2
0
0
6
t
o
th
e
p
re
se
n
t,
h
e
h
a
s
b
e
e
n
a
p
ro
f
e
ss
o
r
in
th
e
S
c
h
o
o
l
o
f
Co
p
u
ter
In
f
o
rm
a
ti
o
n
&
Co
m
m
u
n
ica
ti
o
n
En
g
ii
n
e
e
rin
g
.
Hi
s
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
Im
a
g
e
P
r
o
c
e
ss
in
g
,
Co
m
p
u
ter
V
isi
o
n
,
a
n
d
M
u
lt
im
e
d
ia.
Evaluation Warning : The document was created with Spire.PDF for Python.