T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
23
,
No
.
4
,
A
u
g
u
s
t
20
25
,
p
p
.
9
7
6
~
9
8
5
I
SS
N:
1693
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
23
i
4
.
26856
976
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Enha
ncing
real
is
m
i
n han
d
-
dra
w
n
hu
m
a
n s
k
e
tches
t
hro
ug
h
co
ndition
a
l gener
a
tive a
dv
ersa
ria
l net
w
o
rk
I
m
ra
n Ulla
K
ha
n
1,
2
,
Depa
Ra
m
a
cha
nd
ra
ia
h
K
u
m
a
r
Ra
j
a
1
1
S
c
h
o
o
l
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
R
EV
A
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
S
E
,
S
r
i
K
r
i
sh
n
a
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
1
7
,
2
0
2
4
R
ev
i
s
ed
Ma
y
1
9
,
2
0
2
5
A
cc
ep
ted
Ma
y
2
6
,
2
0
2
5
T
h
is
re
se
a
rc
h
f
o
c
u
se
s
o
n
e
n
h
a
n
c
in
g
th
e
re
a
li
sm
o
f
h
a
n
d
d
ra
w
n
h
u
m
a
n
sk
e
tch
e
s
th
ro
u
g
h
th
e
u
se
o
f
c
o
n
d
it
io
n
a
l
g
e
n
e
ra
ti
v
e
a
d
v
e
rsa
rial
n
e
t
w
o
rk
s
(c
GA
N)
.
A
d
d
re
ss
in
g
th
e
c
h
a
ll
e
n
g
e
o
f
tra
n
sla
ti
n
g
ru
d
im
e
n
tar
y
sk
e
tch
e
s
in
to
h
ig
h
-
f
id
e
li
t
y
i
m
a
g
e
s,
b
y
le
v
e
r
a
g
in
g
th
e
c
a
p
a
b
il
it
y
o
f
d
e
e
p
lea
rn
in
g
a
lg
o
rit
h
m
s
su
c
h
a
s
c
GA
Ns
.
T
h
is
is
p
a
rti
c
u
larly
sig
n
if
ica
n
t
f
o
r
a
p
p
l
ica
ti
o
n
s
in
law
e
n
fo
rc
e
m
e
n
t,
w
h
e
re
a
c
c
u
ra
te
f
a
c
ial
re
c
o
n
stru
c
ti
o
n
f
ro
m
e
y
e
w
it
n
e
ss
sk
e
tch
e
s is
c
r
u
c
ial.
Ou
r
re
se
a
rc
h
u
ti
li
z
e
s
t
h
e
C
h
in
e
se
U
n
iv
e
rsit
y
o
f
Ha
n
g
Ko
n
g
F
a
c
e
S
k
e
tch
e
s
(
CUFS
)
d
a
tas
e
t,
a
p
a
ired
d
a
tas
e
t
o
f
h
a
n
d
d
ra
w
n
h
u
m
a
n
f
a
c
e
s
s
k
e
tch
e
s
a
n
d
th
e
ir
c
o
rre
sp
o
n
d
in
g
re
a
li
stic
im
a
g
e
s
to
train
th
e
c
G
A
N
m
o
d
e
l.
G
e
n
e
ra
to
r
n
e
tw
o
rk
p
ro
d
u
c
e
s
re
a
li
stic
im
a
g
e
s
b
a
se
d
o
n
i
n
p
u
t
sk
e
tch
e
s,
w
h
e
re
a
s
d
isc
rim
in
a
to
r
n
e
tw
o
rk
e
v
a
lu
a
tes
a
u
th
e
n
ti
c
it
y
o
f
th
e
se
g
e
n
e
ra
ted
i
m
a
g
e
s
c
o
m
p
a
re
d
to
th
e
re
a
l
o
n
e
s.
T
h
e
st
u
d
y
in
v
o
lv
e
s
c
a
re
fu
l
p
re
p
ro
c
e
ss
in
g
o
f
th
e
d
a
tas
e
t,
in
c
lu
d
i
n
g
n
o
rm
a
li
z
a
ti
o
n
a
n
d
a
u
g
m
e
n
tatio
n
,
to
e
n
su
re
o
p
ti
m
a
l
train
in
g
c
o
n
d
i
ti
o
n
s.
T
h
e
m
o
d
e
l
p
e
rf
o
r
m
a
n
c
e
a
ss
e
ss
e
d
th
ro
u
g
h
b
o
th
q
u
a
n
ti
tat
iv
e
m
e
tri
c
s,
su
c
h
a
s
f
re
c
h
e
t
in
c
e
p
ti
o
n
d
istan
c
e
(F
ID),
a
n
d
q
u
a
li
tativ
e
e
v
a
lu
a
ti
o
n
s,
i
n
c
lu
d
in
g
v
isu
a
l
in
sp
e
c
ti
o
n
o
f
g
e
n
e
ra
ted
ima
g
e
s.
T
h
e
p
o
ten
ti
a
l
a
p
p
li
c
a
ti
o
n
s
o
f
th
i
s
re
se
a
rc
h
e
x
ten
d
to
v
a
rio
u
s
f
ield
s,
s
u
c
h
a
s
a
g
e
n
c
ies
o
f
law
e
n
f
o
rc
e
m
e
n
t
f
o
r
f
in
d
i
n
g
su
sp
e
c
ts
a
n
d
l
o
c
a
ti
n
g
m
issin
g
p
e
rso
n
s.
F
u
tu
re
w
o
rk
e
x
p
lo
rin
g
a
d
v
a
n
c
e
d
tec
h
n
iq
u
e
s
f
o
r
f
u
rth
e
r
re
a
li
s
m
,
a
n
d
e
v
a
lu
a
ti
n
g
th
e
m
o
d
e
l
’
s
p
e
rf
o
r
m
a
n
c
e
a
c
ro
ss
d
iv
e
rse
d
a
tas
e
ts.
K
ey
w
o
r
d
s
:
C
o
n
d
itio
n
a
l
g
e
n
er
ativ
e
ad
v
er
s
ar
ial
n
et
w
o
r
k
Fre
ch
et
i
n
ce
p
tio
n
d
is
ta
n
ce
Han
d
d
r
a
w
n
h
u
m
an
s
k
etc
h
es
L
a
w
e
n
f
o
r
ce
m
en
t a
p
p
licatio
n
s
R
ea
lis
t
ic
i
m
a
g
e
g
en
er
atio
n
Sk
etc
h
-
to
-
i
m
a
g
e
tr
an
s
latio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
I
m
r
an
Ulla
K
h
a
n
Sch
o
o
l o
f
C
o
m
p
u
ter
Scie
n
ce
,
R
E
V
A
U
n
iv
er
s
it
y
B
an
g
alo
r
e,
I
n
d
ia
E
m
ail:
i
m
r
an
1
6
1
9
8
4
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
A
r
ti
f
icial
in
telli
g
en
ce
(
A
I
)
h
as
s
i
g
n
i
f
ica
n
tl
y
tr
a
n
s
f
o
r
m
ed
v
ar
io
u
s
f
ield
s
,
i
n
cl
u
d
i
n
g
f
o
r
en
s
ic
in
v
e
s
ti
g
atio
n
s
an
d
d
ig
ital
m
ed
ia,
b
y
e
n
h
a
n
ci
n
g
t
h
e
ab
ilit
y
to
an
al
y
ze
a
n
d
g
en
er
ate
i
m
a
g
es.
On
e
o
f
th
e
cr
itica
l
ch
alle
n
g
e
s
in
la
w
en
f
o
r
ce
m
e
n
t
is
id
en
ti
f
y
i
n
g
i
n
d
iv
id
u
al
s
b
ased
o
n
e
y
e
w
it
n
es
s
-
p
r
o
v
id
ed
s
k
etch
e
s
,
esp
ec
iall
y
w
h
e
n
n
o
p
r
io
r
d
ata
is
av
ailab
le.
I
n
cr
ea
s
in
g
th
e
u
s
ag
e
o
f
m
o
b
ile
d
ev
ices
an
d
in
ter
n
et
s
k
e
tch
es
h
a
v
e
b
ec
o
m
e
m
o
r
e
p
o
p
u
lar
w
a
y
to
s
ea
r
ch
a
n
atu
r
al
i
m
a
g
e.
Sk
etc
h
b
ased
im
ag
e
r
etr
ie
v
al
te
ch
n
iq
u
e
u
s
ed
b
y
f
o
r
en
s
ic
a
g
en
c
ie
s
to
ass
is
t
in
id
e
n
ti
f
y
in
g
a
s
u
s
p
ec
t
p
er
s
o
n
in
v
o
lv
ed
i
n
cr
i
m
in
al
ac
tiv
it
ies
w
h
en
th
er
e
i
s
n
o
p
r
io
r
d
ata
av
ailab
le
ab
o
u
t
th
at
p
er
s
o
n
[
1
]
.
C
o
m
p
o
s
ite
o
f
a
s
u
s
p
ec
ted
is
cr
ea
ted
w
i
th
th
e
e
y
e
w
it
n
es
s
b
y
f
o
r
en
s
ic
a
r
tis
t
an
d
au
t
h
o
r
ities
d
is
s
e
m
in
ate
s
t
h
e
co
m
p
o
s
ite
i
m
ag
e
w
it
h
t
h
e
h
o
p
e
s
o
m
eo
n
e
w
i
ll
r
ec
o
g
n
ize
an
d
p
r
o
v
id
es
s
o
m
e
p
er
ti
n
e
n
t
in
f
o
r
m
atio
n
[
2
]
.
W
ith
th
e
in
c
r
ea
s
e
in
cr
i
m
e
ac
tiv
ities
d
a
y
b
y
d
a
y
a
n
d
in
v
o
l
v
e
m
en
t
o
f
n
e
w
p
er
s
o
n
lead
s
a
ch
alle
n
g
i
n
g
j
o
b
f
o
r
t
h
e
co
p
s
to
tr
ac
e
an
d
id
en
t
if
y
th
e
m
.
S
k
etc
h
es
p
la
y
s
a
u
s
e
f
u
l
l
r
o
le
in
t
h
e
ca
s
e,
b
u
t
d
u
e
to
lack
o
f
d
if
f
er
e
n
ce
b
et
w
ee
n
s
k
etc
h
e
s
an
d
r
ea
l
lif
e
im
a
g
es
an
d
also
th
e
less
o
r
lack
o
f
k
n
o
w
led
g
e
ab
o
u
t
p
s
y
c
h
o
lo
g
ical
w
a
y
s
o
f
g
en
er
at
in
g
s
k
etch
e
s
id
en
ti
f
y
in
g
a
cr
i
m
i
n
al
t
h
r
o
u
g
h
s
k
etch
es
h
as
m
ad
e
a
ch
allen
g
i
n
g
j
o
b
w
ith
tr
ad
itio
n
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
E
n
h
a
n
cin
g
r
ea
lis
m
in
h
a
n
d
-
d
r
a
w
n
h
u
ma
n
s
ke
tch
es th
r
o
u
g
h
co
n
d
itio
n
a
l g
en
era
tive
…
(
I
mra
n
Ulla
K
h
a
n
)
977
m
et
h
o
d
s
.
Ou
r
ap
p
r
o
ac
h
to
s
o
lv
e
t
h
e
is
s
u
e
i
s
b
y
u
s
i
n
g
d
ee
p
lear
n
in
g
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
et
w
o
r
k
(
C
NN)
alg
o
r
ith
m
[
3
]
th
at
g
et
s
tr
ain
ed
w
it
h
lar
g
e
d
ataset
an
d
m
o
r
e
f
ea
tu
r
es
o
f
an
i
m
a
g
e,
s
o
th
at
alg
o
r
ith
m
ca
n
ac
cu
r
atel
y
id
en
ti
f
y
a
h
u
m
a
n
f
ac
e
an
d
ex
tr
ac
ts
its
f
ea
t
u
r
es.
T
h
is
co
u
ld
p
o
t
en
tiall
y
b
e
u
s
e
f
u
l
in
la
w
e
n
f
o
r
ce
m
en
t
an
d
f
o
r
en
s
i
c
in
v
e
s
ti
g
atio
n
co
n
tex
t
s
,
w
h
er
e
it
m
a
y
b
e
n
ec
ess
ar
y
to
q
u
ick
l
y
a
n
d
ac
cu
r
atel
y
r
ec
o
g
n
izi
n
g
p
er
s
o
n
b
ased
o
n
s
k
etc
h
es
o
r
o
th
er
v
is
u
al
r
ep
r
ese
n
tat
io
n
s
[
4
]
.
R
ec
en
t
y
ea
r
s
th
e
ar
ea
o
f
co
m
p
u
ter
v
i
s
io
n
h
a
s
m
ad
e
r
e
m
ar
k
ab
l
e
s
tr
id
es,
lar
g
el
y
d
u
e
to
th
e
a
d
v
en
t
an
d
ev
o
l
u
tio
n
o
f
d
ee
p
lear
n
in
g
tech
n
iq
u
e
s
.
A
m
o
n
g
th
ese,
g
en
er
ati
v
e
ad
v
er
s
ar
ial
n
e
t
w
o
r
k
s
(
G
A
Ns)
[
5
]
,
[
6
]
h
av
e
e
m
er
g
ed
a
s
o
n
e
o
f
th
e
m
o
s
t
p
r
o
m
is
in
g
ap
p
r
o
a
ch
es
f
o
r
g
e
n
er
ati
n
g
h
ig
h
-
f
id
ilt
y
s
y
n
t
h
etic
d
ata.
T
h
is
r
esear
ch
f
o
c
u
s
e
s
o
n
h
ar
n
es
s
in
g
th
e
p
o
ten
tial
o
f
co
n
d
itio
n
al
g
en
er
ati
v
e
ad
v
er
s
ar
ial
n
et
w
o
r
k
s
(
cG
A
N
s
)
to
en
h
an
ce
th
e
r
ea
lis
m
o
f
h
an
d
-
d
r
a
w
n
h
u
m
a
n
s
k
etc
h
e
s
[
7
]
.
T
h
e
p
r
im
ar
y
o
b
j
ec
tiv
e
is
to
tr
an
s
f
o
r
m
r
u
d
i
m
en
tar
y
s
k
etch
e
s
in
to
h
i
g
h
-
f
id
elit
y
,
r
ea
lis
tic
i
m
ag
e
s
,
lev
er
ag
i
n
g
t
h
e
s
o
p
h
is
tica
ted
p
o
ten
tialiti
es
o
f
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
s
u
c
h
as
cG
A
N
s
,
w
h
ich
co
n
d
itio
n
t
h
e
g
e
n
er
atio
n
p
r
o
ce
s
s
o
n
s
p
ec
if
ic
in
p
u
t d
ata.
T
r
a
d
itio
n
al
m
et
h
o
d
s
o
f
co
n
v
er
tin
g
s
k
e
tch
e
s
to
r
ea
lis
tic
i
m
a
g
es
h
av
e
b
ee
n
li
m
ited
b
y
th
e
ir
r
elian
ce
o
n
m
an
u
al
tec
h
n
iq
u
es
a
n
d
th
e
la
ck
o
f
s
o
p
h
i
s
ticat
ed
alg
o
r
it
h
m
s
ca
p
ab
le
o
f
ca
p
tu
r
in
g
t
h
e
i
n
tr
icac
ies
o
f
h
u
m
an
f
ea
t
u
r
es
[
8
]
,
[
9
]
.
T
o
ad
d
r
ess
th
ese
li
m
ita
tio
n
s
,
d
ee
p
l
ea
r
n
in
g
tech
n
iq
u
es,
s
p
ec
iall
y
G
A
N
s
h
av
e
e
m
er
g
ed
as
a
p
r
o
m
i
s
in
g
s
o
lu
tio
n
.
cG
A
N
s
ex
ten
d
th
e
ca
p
ab
ilit
ies
o
f
tr
ad
itio
n
al
G
A
N
s
b
y
co
n
d
itio
n
i
n
g
t
h
e
g
en
er
atio
n
p
r
o
ce
s
s
o
n
in
p
u
t
s
k
etch
e
s
,
en
s
u
r
in
g
m
o
r
e
r
ea
lis
tic
im
a
g
e
s
y
n
t
h
es
is
.
T
h
is
r
esear
ch
f
o
cu
s
e
s
o
n
lev
er
ag
i
n
g
cG
A
N
s
to
b
o
o
s
t
th
e
r
ea
lis
m
o
f
h
a
nd
-
d
r
a
w
n
h
u
m
an
s
k
etc
h
e
s
,
th
er
eb
y
i
m
p
r
o
v
i
n
g
ac
c
u
r
ac
y
an
d
e
f
f
ec
tiv
e
n
es
s
o
f
s
k
etc
h
-
b
ased
f
ac
e
r
ec
o
g
n
itio
n
.
I
n
t
h
is
s
t
u
d
y
,
w
e
u
t
ilize
t
h
e
C
h
in
e
s
e
U
n
i
v
er
s
it
y
o
f
Ha
n
g
Ko
n
g
Face
Sk
etc
h
e
s
(
C
U
FS
)
d
at
aset,
w
h
ich
co
n
s
is
ts
o
f
p
air
ed
h
an
d
-
d
r
a
w
n
s
k
etc
h
es
a
n
d
th
eir
m
ap
p
e
d
r
ea
l
im
ag
e
s
,
to
tr
ain
a
cGA
N
m
o
d
el
[
1
0
]
.
T
h
e
ev
alu
a
tio
n
o
f
o
u
r
ap
p
r
o
ac
h
is
co
n
d
u
cted
u
s
in
g
f
r
ec
h
et
i
n
ce
p
ti
o
n
d
is
tan
ce
(
FID
)
[
1
1
]
,
w
h
ich
m
ea
s
u
r
es
t
h
e
q
u
ali
t
y
o
f
g
e
n
er
ated
i
m
ag
e
s
.
O
u
r
wo
r
k
ai
m
s
to
b
r
id
g
e
t
h
e
g
ap
b
et
w
ee
n
f
o
r
en
s
ic
s
k
etch
e
s
a
n
d
r
ea
l
-
w
o
r
ld
f
ac
ial
r
ec
o
g
n
itio
n
b
y
d
e
v
elo
p
in
g
a
n
A
I
-
d
r
iv
e
n
m
o
d
el
th
at
ca
n
ac
c
u
r
atel
y
r
ec
o
n
s
tr
u
ct
h
u
m
an
f
ac
e
s
f
r
o
m
h
an
d
-
d
r
a
w
n
s
k
etc
h
es.
T
h
i
s
ad
v
a
n
ce
m
en
t
h
o
ld
s
s
ig
n
i
f
ica
n
t
p
o
ten
t
ial
i
n
f
o
r
en
s
ic
i
n
v
e
s
ti
g
atio
n
s
,
s
u
s
p
ec
t
id
en
ti
f
icatio
n
,
an
d
d
ig
ital a
r
t a
p
p
licatio
n
s
T
h
is
s
tu
d
y
ai
m
s
to
g
e
n
er
ate
n
e
w
,
h
i
g
h
r
eso
lu
tio
n
h
u
m
an
f
ac
e
i
m
ag
e
s
an
d
i
m
p
r
o
v
e
th
e
q
u
a
li
t
y
o
f
th
e
s
e
i
m
a
g
es
u
s
i
n
g
G
A
N
s
.
Sp
ec
if
ic
all
y
,
t
h
e
s
t
u
d
y
e
m
p
lo
y
s
a
co
m
b
in
atio
n
o
f
d
ee
p
co
n
v
o
lu
t
io
n
a
l
GANs
(
D
C
G
A
N)
an
d
en
h
a
n
ce
d
s
u
p
er
-
r
e
s
o
lu
tio
n
G
A
Ns
(
E
SR
G
A
N)
.
DC
G
AN
u
s
es
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
w
o
r
k
s
to
g
en
er
at
e
i
m
a
g
es
f
r
o
m
r
an
d
o
m
n
o
i
s
e,
wh
ile
E
SR
G
AN
e
n
h
a
n
ce
s
t
h
e
r
e
s
o
lu
tio
n
a
n
d
q
u
alit
y
o
f
t
h
ese
i
m
ag
e
s
.
T
h
e
C
eleb
A
d
ataset,
co
n
tain
in
g
o
v
er
2
5
,
0
0
0
ce
leb
r
ity
f
ac
e
i
m
ag
e
s
,
w
as
u
s
ed
f
o
r
tr
ain
in
g
.
T
h
e
r
es
u
lt
s
s
h
o
w
t
h
at
t
h
e
co
m
b
i
n
ed
ap
p
r
o
ac
h
o
f
DC
G
A
N
a
n
d
E
SR
G
A
N
ef
f
ec
ti
v
el
y
p
r
o
d
u
ce
s
h
i
g
h
q
u
alit
y
h
u
m
a
n
f
ac
e
s
,
w
it
h
i
m
p
r
o
v
e
m
en
ts
m
ea
s
u
r
ed
u
s
i
n
g
t
h
e
s
tr
u
ct
u
r
al
s
i
m
ilar
it
y
i
n
d
ex
(
SS
I
M)
.
Desp
ite
th
e
ad
v
an
ce
m
e
n
ts
,
th
e
s
tu
d
y
n
o
tes
li
m
itatio
n
s
in
g
e
n
er
ati
n
g
h
i
g
h
f
id
eli
t
y
i
m
a
g
es
a
n
d
ca
p
tu
r
in
g
i
n
tr
icate
d
et
ails
,
in
d
icati
n
g
p
o
ten
tial
f
o
r
f
u
r
th
er
en
h
an
ce
m
e
n
t
w
it
h
e
x
ten
d
ed
tr
ain
in
g
an
d
f
in
e
-
t
u
n
i
n
g
o
f
m
o
d
el
p
ar
a
m
eter
s
[
1
2
]
.
I
n
t
h
is
r
e
s
ea
r
ch
w
o
r
k
h
as
d
e
v
el
o
p
a
h
ig
h
f
id
elit
y
f
ac
e
g
e
n
er
ati
o
n
m
o
d
el
u
s
in
g
St
y
leG
A
N.
T
h
is
r
esear
ch
u
tili
ze
s
p
u
b
licl
y
o
w
n
ed
d
atas
ets,
s
p
ec
i
f
icall
y
t
h
e
Fli
c
k
r
H
Q
Data
s
et
a
n
d
t
h
e
Me
t
f
ac
es
Data
s
et.
T
h
e
p
r
i
m
ar
y
o
b
j
ec
tiv
e
is
to
g
en
er
ate
d
i
v
er
s
e
an
d
r
ea
lis
t
ic
f
ac
ia
l
i
m
ag
e
s
f
r
o
m
te
x
t
u
al
d
escr
ip
tio
n
s
.
H
er
e
th
e
y
h
a
v
e
u
s
ed
St
y
le
G
A
N,
tr
ain
ed
o
n
a
lar
g
e
d
ataset
o
f
h
u
m
a
n
f
ac
e
s
to
en
s
u
r
e
h
ig
h
q
u
al
it
y
o
u
tp
u
t
s
.
T
h
e
p
o
ten
tial
ap
p
licatio
n
s
o
f
th
i
s
m
o
d
el
s
p
a
n
v
ar
io
u
s
f
i
eld
s
,
in
cl
u
d
in
g
cr
i
m
in
al
in
v
es
tig
atio
n
s
,
f
ac
e
r
ec
o
g
n
i
tio
n
s
y
s
te
m
a
u
g
m
e
n
tatio
n
,
co
m
p
u
ter
g
r
ap
h
ics,
a
n
d
en
ter
t
ain
m
en
t.
Ho
w
e
v
er
,
t
h
er
e
is
a
n
ee
d
f
o
r
ex
ten
s
iv
e
co
m
p
u
tati
o
n
al
r
eso
u
r
ce
s
a
n
d
p
o
ten
tial b
iases
i
n
th
e
tr
ai
n
i
n
g
d
ataset
th
at
co
u
ld
a
f
f
ec
t t
h
e
g
en
er
ated
i
m
a
g
es d
iv
er
s
it
y
a
n
d
ac
cu
r
ac
y
[
1
3
]
T
h
is
r
esear
ch
p
ap
er
Ko
v
ar
t
h
a
n
an
an
d
K
u
m
ar
asi
n
g
h
e
[
1
4
]
h
as
s
h
o
w
n
h
o
w
to
e
n
h
an
ce
th
e
r
ea
lis
m
in
s
k
etc
h
-
to
-
i
m
a
g
e
tr
an
s
latio
n
u
s
in
g
cG
ANs.
T
h
e
m
e
th
o
d
o
lo
g
y
in
v
o
l
v
es
u
s
in
g
a
cG
A
N
m
o
d
e
l,
w
h
ich
co
m
b
i
n
es
a
g
en
er
ato
r
an
d
a
d
is
cr
im
i
n
at
o
r
in
an
ad
v
er
s
ar
ial
s
etu
p
to
p
r
o
d
u
ce
r
ea
lis
tic
i
m
ag
e
s
f
r
o
m
i
n
p
u
t
s
k
etc
h
es.
T
h
e
d
atas
et
u
s
ed
f
o
r
tr
ain
i
n
g
a
n
d
test
in
g
t
h
e
m
o
d
el
is
th
e
“
An
i
m
e
Sk
etch
C
o
lo
r
izatio
n
P
air
”
d
ataset
f
r
o
m
Ka
g
g
le,
co
n
s
is
tin
g
o
f
o
v
er
1
5
,
0
0
0
p
ai
r
s
o
f
a
n
i
m
e
s
k
etc
h
es
a
n
d
t
h
e
ir
co
r
r
esp
o
n
d
in
g
co
lo
r
ized
v
e
r
s
io
n
s
.
C
h
a
llen
g
es
id
en
ti
f
ied
in
th
e
r
esear
c
h
is
t
h
e
co
m
p
u
tat
io
n
al
co
m
p
le
x
it
y
an
d
th
e
p
o
ten
tial
f
o
r
o
v
er
f
itti
n
g
d
u
e
to
th
e
h
i
g
h
d
i
m
en
s
io
n
al
it
y
o
f
th
e
d
ata.
T
h
is
r
esear
ch
ai
m
s
to
d
ev
elo
p
an
ad
v
an
ce
d
GAN
m
o
d
el
f
o
r
g
en
er
ati
n
g
r
ea
lis
tic
co
lo
u
r
i
m
ag
es
f
r
o
m
h
u
m
a
n
f
ac
e
s
k
etc
h
es.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el,
an
atte
n
tio
n
-
b
ased
co
n
tex
t
u
al
G
A
N,
le
v
er
ag
es
th
e
p
o
w
er
o
f
R
esNet
-
5
0
f
o
r
h
ig
h
-
lev
el
f
ac
ia
l f
ea
t
u
r
e
ex
tr
ac
tio
n
.
T
h
is
G
A
N
u
s
es a
n
o
v
el
s
el
f
-
at
ten
t
io
n
m
ec
h
a
n
i
s
m
,
w
h
ich
a
llo
w
s
t
h
e
g
en
er
ato
r
to
f
o
cu
s
o
n
cr
u
cial
ele
m
e
n
ts
o
f
th
e
s
k
etc
h
es,
p
r
o
d
u
cin
g
h
i
g
h
-
q
u
alit
y
a
n
d
d
etailed
im
a
g
es.
D
u
r
in
g
tr
a
in
i
n
g
,
a
co
m
b
in
ed
lo
s
s
f
u
n
ct
io
n
,
in
co
r
p
o
r
atin
g
b
o
t
h
p
ix
el
a
n
d
co
n
te
x
tu
a
l
lo
s
s
e
s
,
en
s
u
r
e
s
t
h
e
g
e
n
er
ated
i
m
ag
e
s
clo
s
el
y
m
atc
h
th
e
g
r
o
u
n
d
tr
u
th
[
1
5
]
.
T
h
is
p
ap
er
is
s
tr
u
ctu
r
ed
as
f
o
l
lo
w
s
,
s
ec
tio
n
2
de
tails
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
lo
g
y
,
m
o
d
el
ar
ch
itect
u
r
e
an
d
tr
ain
i
n
g
s
tr
ateg
y
.
Sectio
n
3
p
r
esen
ts
e
x
p
er
i
m
e
n
tal
r
e
s
u
l
t
s
,
d
ataset
p
r
ep
r
o
ce
s
s
in
g
f
o
llo
w
ed
b
y
a
d
is
c
u
s
s
io
n
o
n
k
e
y
f
i
n
d
in
g
s
.
Fin
a
ll
y
s
ec
tio
n
4
c
o
n
clu
d
es t
h
e
s
t
u
d
y
w
i
th
f
u
tu
r
e
r
esear
c
h
d
ir
ec
tio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
23
,
No
.
4
,
A
u
g
u
s
t
20
25
:
9
7
6
-
9
85
978
2.
M
E
T
H
O
D
As
m
en
t
io
n
ed
i
n
Fi
g
u
r
e
1
t
h
e
p
r
o
p
o
s
ed
r
esea
r
ch
s
y
s
te
m
a
i
m
s
t
o
in
cr
ea
s
e
r
ea
lis
m
o
f
h
an
d
-
d
r
aw
n
h
u
m
a
n
s
k
etc
h
es
b
y
h
ar
n
e
s
s
i
n
g
t
h
e
ca
p
ab
ilit
ies
o
f
cG
ANs
.
c
G
A
N
i
s
an
ex
te
n
s
io
n
o
f
G
AN
th
at
i
n
co
r
p
o
r
ate
c
o
n
d
itio
n
al
in
f
o
r
m
atio
n
,
s
u
ch
as
in
p
u
t
i
m
ag
es,
to
g
u
id
e
t
h
e
g
en
er
atio
n
p
r
o
ce
s
s
.
I
n
t
h
is
r
e
s
ea
r
ch
,
cG
A
Ns
le
v
er
ag
e
p
air
ed
s
k
etc
h
-
i
m
ag
e
d
ata
to
g
e
n
er
ate
r
ea
lis
tic
h
u
m
a
n
f
ac
es
f
r
o
m
h
an
d
-
d
r
a
w
n
s
k
etch
e
s
b
y
lea
r
n
in
g
th
e
m
ap
p
in
g
b
et
w
ee
n
th
e
t
w
o
d
o
m
ai
n
s
.
Fig
u
r
e
1
.
I
m
a
g
e
co
n
v
er
s
io
n
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
s
tar
t
s
w
it
h
an
i
n
p
u
t
s
k
etc
h
,
w
h
ich
u
n
d
er
g
o
es
p
r
ep
r
o
ce
s
s
in
g
s
te
p
s
s
u
c
h
as
n
o
r
m
aliza
t
io
n
an
d
au
g
m
e
n
ta
tio
n
to
en
s
u
r
e
o
p
ti
m
al
tr
ai
n
in
g
co
n
d
it
io
n
s
as
d
escr
ib
ed
in
Fig
u
r
e
2
.
T
h
e
p
r
ep
r
o
ce
s
s
ed
s
k
etc
h
e
s
ar
e
t
h
en
f
ed
i
n
to
t
h
e
g
e
n
er
ato
r
n
et
w
o
r
k
(
G)
w
h
ich
is
co
m
p
o
s
ed
o
f
a
n
e
n
co
d
er
b
lo
ck
an
d
a
d
ec
o
d
e
r
b
lo
ck
.
T
h
e
en
co
d
er
co
n
v
er
ts
t
h
e
s
k
etc
h
in
to
a
laten
t
r
ep
r
esen
tatio
n
,
ca
p
tu
r
i
n
g
t
h
e
ess
en
tial
f
ea
t
u
r
es
r
eq
u
ir
ed
f
o
r
r
ea
lis
tic
i
m
a
g
e
g
e
n
er
atio
n
[
1
6
]
,
[
1
7
]
.
T
h
is
laten
t
r
ep
r
esen
tatio
n
is
t
h
e
n
p
ass
ed
t
h
r
o
u
g
h
t
h
e
d
ec
o
d
er
t
o
p
r
o
d
u
ce
a
h
ig
h
-
f
id
elit
y
i
m
a
g
e
[
1
8
]
,
[
1
9
]
.
A
s
s
h
o
w
n
b
elo
w
i
n
d
is
cr
i
m
i
n
ato
r
n
et
w
o
r
k
(
D
)
is
u
s
ed
to
ev
al
u
at
e
th
e
au
th
e
n
tici
t
y
o
f
th
e
g
e
n
er
at
ed
i
m
ag
e
s
b
y
co
m
p
ar
i
n
g
th
e
m
to
r
ea
l
i
m
ag
e
s
,
lear
n
i
n
g
to
d
if
f
er
en
tia
te
b
et
w
ee
n
r
ea
l a
n
d
f
ak
e
g
en
er
a
ted
d
ata.
Fig
u
r
e
2
.
S
y
s
te
m
ar
ch
itect
u
r
e
cGAN
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
E
n
h
a
n
cin
g
r
ea
lis
m
in
h
a
n
d
-
d
r
a
w
n
h
u
ma
n
s
ke
tch
es th
r
o
u
g
h
co
n
d
itio
n
a
l g
en
era
tive
…
(
I
mra
n
Ulla
K
h
a
n
)
979
2
.
1
.
G
ener
a
t
o
r
net
w
o
rk
T
h
e
g
en
er
ato
r
n
et
w
o
r
k
i
n
o
u
r
p
r
o
j
ec
t u
tili
ze
s
a
U
-
Net
ar
ch
it
ec
tu
r
e
as i
n
Fi
g
u
r
e
3
,
a
w
id
el
y
u
s
ed
C
N
N
f
o
r
i
m
a
g
e
-
to
-
i
m
a
g
e
tr
a
n
s
latio
n
task
s
[
2
0
]
.
U
-
Net
co
n
s
is
ts
o
f
a
n
en
co
d
er
d
ec
o
d
er
w
it
h
s
k
ip
co
n
n
ec
t
io
n
s
,
i
t
allo
w
s
to
ca
p
tu
r
e
b
o
th
g
lo
b
al
co
n
tex
t
an
d
f
in
e
-
g
r
ai
n
ed
d
etails
ef
f
ec
tiv
el
y
.
T
h
e
en
co
d
er
ex
tr
ac
ts
h
ier
ar
ch
ical
f
ea
t
u
r
es
f
r
o
m
g
i
v
en
s
k
etc
h
,
w
h
ile
t
h
e
d
ec
o
d
er
r
ec
o
n
s
tr
u
cts
a
h
i
g
h
-
r
eso
lu
tio
n
r
ea
lis
tic
f
ac
e
i
m
a
g
e
.
Sk
ip
co
n
n
ec
tio
n
s
b
r
id
g
e
co
r
r
esp
o
n
d
in
g
la
y
er
s
in
t
h
e
e
n
co
d
er
an
d
d
ec
o
d
er
,
p
r
eser
v
in
g
s
p
atial
i
n
f
o
r
m
ati
o
n
an
d
i
m
p
r
o
v
i
n
g
r
ec
o
n
s
tr
u
ctio
n
q
u
alit
y
.
T
h
is
ar
ch
i
tect
u
r
e
en
h
a
n
ce
s
t
h
e
g
en
e
r
ato
r
’
s
ab
ilit
y
to
p
r
o
d
u
ce
r
ea
l
is
tic
i
m
a
g
es
w
h
ile
m
ai
n
tai
n
in
g
s
tr
u
ct
u
r
al
co
n
s
i
s
te
n
c
y
w
i
th
t
h
e
i
n
p
u
t s
k
etc
h
.
Fig
u
r
e
3
. U
-
Net
ar
ch
i
tectu
r
e
(
g
en
er
ato
r
n
et
w
o
r
k
)
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
in
v
o
l
v
es
m
i
n
i
m
izi
n
g
t
w
o
k
e
y
lo
s
s
e
s
:
th
e
co
n
d
it
io
n
al
lo
s
s
(
L
1
)
an
d
t
h
e
ad
v
er
s
ar
ial
lo
s
s
.
T
h
e
co
n
d
itio
n
al
lo
s
s
en
s
u
r
es
th
at
th
e
o
u
tp
u
t
i
m
a
g
es
cl
o
s
el
y
r
ese
m
b
le
th
e
d
esi
r
ed
r
ea
lis
tic
i
m
a
g
e
s
,
w
h
ile
ad
v
er
s
ar
ial
lo
s
s
d
r
iv
es
t
h
e
g
en
er
ato
r
to
p
r
o
d
u
ce
i
m
a
g
es
t
h
at
ar
e
i
n
d
is
ti
n
g
u
is
h
ab
le
f
r
o
m
r
ea
l
i
m
a
g
es.
T
h
e
g
en
er
ato
r
b
lo
ck
an
d
d
is
cr
i
m
i
n
ato
r
b
lo
ck
n
et
w
o
r
k
s
ar
e
u
p
d
ated
iter
ativ
el
y
,
w
it
h
ea
ch
it
er
atio
n
r
ef
in
in
g
th
e
m
o
d
el
’
s
w
e
ig
h
t
s
th
r
o
u
g
h
th
e
o
p
tim
izer
.
T
h
is
iter
ati
v
e
tr
a
in
i
n
g
co
n
tin
u
e
s
u
n
til
t
h
e
g
e
n
er
ato
r
co
n
s
is
ten
t
l
y
p
r
o
d
u
ce
s
r
ea
lis
tic
i
m
a
g
es
f
r
o
m
s
k
etc
h
es.
2
.
1
.
1
.
Adv
er
s
a
ria
l
lo
s
s
f
o
r
g
e
nera
t
o
r
(
G
)
T
h
e
ad
v
er
s
ar
ial
lo
s
s
m
o
ti
v
ates
g
en
er
ato
r
to
p
r
o
d
u
ce
r
ea
lis
tic
im
ag
e
s
t
h
a
t
ca
n
m
i
s
lead
th
e
d
is
cr
i
m
i
n
ato
r
as
in
(
1
)
.
I
t
is
f
o
r
m
u
lated
as
a
m
i
n
i
m
izatio
n
p
r
o
b
lem
w
h
er
e
G
tr
ies
to
m
a
x
i
m
ize
th
e
d
is
cr
i
m
in
a
to
r
’
s
class
i
f
icatio
n
er
r
o
r
.
T
h
is
lo
s
s
d
r
iv
es
t
h
e
g
e
n
er
ato
r
to
p
r
o
d
u
ce
h
ig
h
l
y
r
ea
lis
t
ic
i
m
a
g
es
b
y
co
n
t
in
u
o
u
s
l
y
i
m
p
r
o
v
in
g
its
o
u
tp
u
ts
ag
a
in
s
t t
h
e
d
is
cr
i
m
i
n
ato
r
’
s
e
v
al
u
atio
n
s
.
ℒ
(
)
=
∼
(
)
[
l
ogD
(
x
)
]
+
∼
(
)
[
l
og
(
1
−
(
(
)
)
)
]
(
1
)
ℒ
(
)
is
ad
v
er
s
ar
ial
lo
s
s
f
o
r
th
e
g
e
n
er
ato
r
G.
I
t
q
u
an
ti
f
ies
h
o
w
w
el
l G
f
o
o
ls
th
e
d
is
cr
i
m
i
n
ato
r
D.
E[.]
is
ex
p
ec
ted
v
al
u
e,
x
i
s
s
a
m
p
le
d
f
r
o
m
th
e
tr
u
e
d
ata
d
is
tr
ib
u
tio
n
D(
x
)
is
t
h
e
d
is
cr
i
m
i
n
ato
r
’
s
o
u
tp
u
t
,
z
is
s
a
m
p
led
f
r
o
m
th
e
la
ten
t
s
p
ac
e
d
is
tr
ib
u
tio
n
P
z
(
z)
G(
z)
:
T
h
e
g
en
er
ato
r
’
s
o
u
tp
u
t
w
h
en
g
iv
en
la
ten
t
v
ec
to
r
z,
w
h
ic
h
is
a
f
ak
e
s
a
m
p
le
2
.
1
.
2
.
Co
nd
it
io
na
l
lo
s
s
(
L
1
l
o
s
s
)
T
h
is
lo
s
s
en
s
u
r
es
t
h
at
th
e
g
e
n
er
ated
im
a
g
es
ar
e
s
i
m
ilar
to
t
h
e
ac
tu
al
i
m
a
g
es
i
n
th
e
d
atase
t
en
s
u
r
in
g
s
tr
u
ct
u
r
al
co
n
s
i
s
ten
c
y
.
I
t
h
elp
s
th
e
g
e
n
er
ato
r
f
o
cu
s
o
n
p
r
eser
v
i
n
g
f
i
n
e
d
etails
b
y
m
i
n
i
m
izi
n
g
t
h
e
ab
s
o
lu
te
d
if
f
er
e
n
ce
b
et
w
ee
n
co
r
r
esp
o
n
d
in
g
p
ix
e
ls
as i
n
(
2
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
23
,
No
.
4
,
A
u
g
u
s
t
20
25
:
9
7
6
-
9
85
980
ℒ
1
(
)
=
,
∼
(
,
)
[
|
|
−
(
)
|
|
1
]
(
2
)
T
h
e
g
en
er
ato
r
’
s
to
tal
lo
s
s
co
m
b
in
es
b
o
th
ad
v
er
s
ar
ial
an
d
co
n
d
itio
n
al
lo
s
s
es
a
s
in
(
3
)
,
en
s
u
r
i
n
g
b
o
th
r
ea
lis
m
a
n
d
s
tr
u
ct
u
r
al
ac
c
u
r
ac
y
.
T
h
e
ad
v
er
s
ar
ial
lo
s
s
d
r
iv
es
th
e
g
e
n
er
ato
r
to
cr
ea
te
i
m
a
g
es
in
d
i
s
ti
n
g
u
i
s
h
a
b
le
f
r
o
m
r
ea
l
o
n
es,
w
h
ile
t
h
e
L
1
lo
s
s
p
r
eser
v
es
f
i
n
e
-
g
r
ai
n
ed
d
etails.
T
h
is
co
m
b
in
ed
o
b
j
ec
tiv
e
h
elp
s
ac
h
ie
v
e
h
i
g
h
-
q
u
alit
y
an
d
r
ea
lis
tic
s
k
etc
h
-
to
-
i
m
ag
e
tr
a
n
s
latio
n
s
.
ℒ
=
ℒ
+
ℒ
1
(
3
)
w
h
er
e
λ
is
a
w
ei
g
h
t f
ac
to
r
b
ala
n
cin
g
t
h
e
t
w
o
lo
s
s
e
s
.
2
.
2
.
Descri
m
ina
t
o
r
net
w
o
rk
T
h
e
P
atch
GAN
d
is
cr
i
m
i
n
ato
r
s
h
o
w
n
i
n
Fig
u
r
e
4
b
r
ea
k
s
th
e
g
en
er
ated
i
m
ag
e
o
r
r
ea
l
im
ag
e
i
n
to
s
m
aller
p
atch
es
a
n
d
ev
al
u
ates
ea
c
h
p
a
tch
f
o
r
r
ea
lis
m
,
r
at
h
er
th
a
n
a
s
s
ess
i
n
g
t
h
e
en
t
ir
e
i
m
a
g
e
at
o
n
ce
[
2
1
]
.
T
h
is
p
atch
-
b
ased
ap
p
r
o
ac
h
en
ab
les
f
i
n
er
-
g
r
ai
n
ed
f
ee
d
b
ac
k
.
T
h
i
s
m
ec
h
an
i
s
m
h
e
lp
s
t
h
e
g
e
n
er
ato
r
t
o
i
m
p
r
o
v
e
lo
ca
lized
d
etails an
d
ca
p
tu
r
e
h
i
g
h
-
f
r
eq
u
en
c
y
f
ea
t
u
r
es
m
o
r
e
e
f
f
ec
t
iv
el
y
.
Fig
u
r
e
4
.
P
atch
GA
N
(
d
is
cr
i
m
i
n
ato
r
)
2
.
2
.
1
.
Dis
cr
i
m
ina
t
o
r
lo
s
s
T
h
e
d
is
cr
im
in
ato
r
tar
g
et
s
to
ac
cu
r
etl
y
cla
s
s
i
f
y
r
ea
l
an
d
p
r
o
d
u
ce
d
im
a
g
e
s
.
T
h
e
d
is
cr
i
m
i
n
ato
r
lo
s
s
m
ea
s
u
r
es
its
ab
ilit
y
to
d
is
tin
g
u
i
s
h
b
et
w
ee
n
r
ea
l
an
d
g
en
e
r
ated
im
a
g
es
as
i
n
(
4
)
.
I
t
is
co
m
p
o
s
ed
o
f
t
w
o
co
m
p
o
n
e
n
t
s
:
o
n
e
t
h
at
p
e
n
aliz
es
m
i
s
clas
s
i
f
y
in
g
r
ea
l
i
m
a
g
es
as
f
a
k
e
a
n
d
an
o
th
er
t
h
at
p
e
n
al
izes
m
is
cla
s
s
i
f
y
in
g
g
en
er
ated
i
m
a
g
es
as
r
ea
l.
B
y
m
i
n
i
m
izi
n
g
th
i
s
lo
s
s
,
th
e
d
is
cr
i
m
i
n
ato
r
i
m
p
r
o
v
es
its
ca
p
ab
ilit
y
to
co
r
r
ec
tly
d
if
f
er
e
n
tiate
b
et
w
ee
n
au
th
e
n
tic
an
d
s
y
n
t
h
esized
i
m
a
g
es,
t
h
er
eb
y
p
u
s
h
i
n
g
t
h
e
g
en
er
ato
r
to
p
r
o
d
u
ce
m
o
r
e
r
ea
lis
ti
c
o
u
tp
u
ts
.
ℒ
=
−
(
∼
(
)
[
l
ogD
(
x
)
]
+
∼
(
)
[
l
og
(
1
−
(
(
)
)
)
]
)
(
4
)
T
h
ese
lo
s
s
es
g
u
id
e
th
e
o
p
ti
m
iz
atio
n
p
r
o
ce
s
s
,
lead
in
g
to
t
h
e
ite
r
ativ
e
i
m
p
r
o
v
e
m
e
n
t
o
f
t
h
e
g
en
er
ato
r
an
d
d
is
cr
i
m
i
n
ato
r
[
2
2
]
.
T
h
e
iter
ati
v
e
p
r
o
ce
s
s
k
ee
p
o
n
r
u
n
n
in
g
u
n
til
th
e
g
en
er
ato
r
co
n
s
i
s
ten
t
l
y
p
r
o
d
u
ce
s
h
ig
h
q
u
alit
y
i
m
a
g
es
t
h
at
ar
e
i
n
d
is
ti
n
g
u
i
s
h
a
b
le
f
r
o
m
r
ea
l
d
atab
ase
i
m
ag
e
s
.
T
h
e
ef
f
ec
ti
v
e
n
ess
o
f
t
h
is
ap
p
r
o
ac
h
is
m
ea
s
u
r
ed
th
r
o
u
g
h
b
o
th
q
u
an
ti
tati
v
e
m
etr
ics,
s
u
c
h
as FI
D,
an
d
q
u
alitativ
e
ev
al
u
atio
n
s
,
lik
e
v
is
u
al
in
s
p
ec
tio
n
o
f
g
en
er
ated
i
m
a
g
e
s
as i
n
(
5
)
.
=
∥
−
∥
2
2
+
(
Σ
+
Σ
−
2
(
Σ
Σ
)
1
2
)
(
5
)
−
μ
r
an
d
Σ
r
,
μ
g
an
d
Σ
g
b
e
th
e
m
e
an
an
d
co
v
ar
ia
n
ce
m
atr
i
x
o
f
r
e
al
an
d
g
en
er
ated
i
m
ag
e
s
.
−
∥
μ
r
−μ
g
∥
2
2
r
e
p
r
esen
ts
th
e
s
q
u
ar
ed
E
u
clid
ea
n
d
is
tan
ce
b
et
w
ee
n
th
e
m
ea
n
s
o
f
th
e
o
r
ig
i
n
al
an
d
f
ak
e
i
m
ag
e
f
ea
t
u
r
e
d
is
tr
ib
u
tio
n
s
.
−
Tr
d
en
o
te
s
t
h
e
tr
ac
e
o
f
t
h
e
m
at
r
ix
,
Σ
r
an
d
Σ
g
ar
e
t
h
e
co
v
ar
ian
ce
m
atr
ice
s
o
f
th
e
r
ea
l
a
n
d
g
e
n
er
ated
i
m
ag
e
s
,
r
esp
ec
tiv
el
y
.
(Σ
r
Σ
g
)
1/2
d
en
o
te
s
t
h
e
m
a
tr
ix
s
q
u
ar
e
r
o
o
t o
f
th
e
p
r
o
d
u
ct
o
f
th
e
co
v
ar
ia
n
ce
m
atr
i
ce
s
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
d
e
m
o
n
s
tr
ate
s
ig
n
if
ican
t
p
o
ten
tial
i
n
tr
an
s
f
o
r
m
i
n
g
r
u
d
i
m
e
n
tar
y
s
k
etc
h
es
[
2
3
]
in
to
d
etailed
,
r
ea
lis
tic
im
a
g
es,
w
ith
ap
p
licatio
n
s
in
ar
ea
s
s
u
ch
as
la
w
en
f
o
r
ce
m
e
n
t,
d
ig
ital
ar
t,
an
d
au
to
m
ated
s
k
etc
h
-
to
-
i
m
ag
e
tr
a
n
s
latio
n
[
2
4
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
E
n
h
a
n
cin
g
r
ea
lis
m
in
h
a
n
d
-
d
r
a
w
n
h
u
ma
n
s
ke
tch
es th
r
o
u
g
h
co
n
d
itio
n
a
l g
en
era
tive
…
(
I
mra
n
Ulla
K
h
a
n
)
981
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
r
esu
lts
o
f
t
h
is
s
tu
d
y
d
e
m
o
n
s
tr
ate
th
e
e
f
f
ec
ti
v
en
e
s
s
o
f
c
GANs
i
n
en
h
an
c
in
g
th
e
r
ea
li
s
m
o
f
h
a
n
d
-
d
r
a
w
n
h
u
m
a
n
s
k
etc
h
e
s
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
w
a
s
tr
ain
ed
o
n
th
e
C
UF
S
d
ataset
a
n
d
ev
alu
ated
u
s
i
n
g
b
o
t
h
q
u
an
tita
tiv
e
m
etr
ics,
s
u
c
h
a
s
FID
,
an
d
q
u
a
litati
v
e
v
is
u
al
as
s
ess
m
e
n
ts
.
T
h
e
s
y
s
te
m
ac
h
ie
v
ed
h
ig
h
p
r
ec
is
io
n
i
n
g
en
er
ati
n
g
lif
e
lik
e
i
m
a
g
es,
w
i
th
m
i
n
i
m
al
ar
ti
f
ac
ts
an
d
i
m
p
r
o
v
ed
s
tr
u
ctu
r
al
ac
c
u
r
ac
y
.
A
d
d
itio
n
all
y
,
tr
ac
k
in
g
p
er
f
o
r
m
a
n
ce
m
e
tr
ics,
in
cl
u
d
in
g
FP
S,
I
D
s
w
itc
h
es,
an
d
m
u
lti
o
b
j
ec
t
tr
ac
k
in
g
ac
cu
r
ac
y
(
M
OT
A
)
,
v
alid
at
ed
th
e
ef
f
icien
c
y
o
f
r
ea
l
-
ti
m
e
f
ac
e
d
etec
tio
n
an
d
tr
ac
k
i
n
g
.
T
h
ese
f
i
n
d
in
g
s
h
ig
h
li
g
h
t
th
e
p
o
ten
t
ial
ap
p
licatio
n
s
o
f
th
e
m
o
d
el
i
n
la
w
e
n
f
o
r
ce
m
en
t,
d
i
g
ital a
r
t,
an
d
au
to
m
ated
s
k
etc
h
-
to
-
i
m
ag
e
tr
a
n
s
latio
n
.
3
.
1
.
Da
t
a
prepro
ce
s
s
ing
I
n
t
h
is
r
e
s
ea
r
ch
w
o
r
k
,
w
e
u
t
ili
ze
th
e
C
UF
S
f
ac
e
s
k
etc
h
d
atas
et,
w
h
ic
h
co
n
tain
s
p
air
ed
g
r
a
y
s
ca
le
h
a
n
d
-
d
r
a
w
n
s
k
etc
h
e
s
an
d
th
e
ir
co
r
r
esp
o
n
d
in
g
r
ea
lis
tic
f
ac
e
i
m
ag
e
s
,
to
tr
ain
o
u
r
cG
A
N.
T
h
e
d
ataset
is
p
r
ep
r
o
ce
s
s
ed
to
en
s
u
r
e
u
n
i
f
o
r
m
it
y
an
d
en
h
an
ce
m
o
d
el
p
er
f
o
r
m
an
ce
.
P
r
e
-
p
r
o
ce
s
s
i
n
g
s
tep
s
i
n
cl
u
d
e
r
esi
zin
g
all
i
m
a
g
es
to
a
co
n
s
is
ten
t
r
eso
l
u
tio
n
(
2
5
6
×
2
5
6
p
ix
els),
n
o
r
m
aliz
in
g
p
ix
e
l
v
a
lu
es
to
t
h
e
r
an
g
e
[
0
,
1
]
,
an
d
co
n
v
er
ti
n
g
s
k
etc
h
es
to
g
r
a
y
s
ca
le.
T
o
im
p
r
o
v
e
m
o
d
el
g
en
er
aliza
tio
n
a
n
d
to
r
ed
u
ce
th
e
r
is
k
o
f
o
v
er
f
itt
in
g
,
d
ata
au
g
m
en
tatio
n
tech
n
iq
u
e
s
s
u
c
h
as
r
a
n
d
o
m
r
o
tatio
n
s
,
f
lip
s
,
an
d
s
h
i
f
t
s
ar
e
ap
p
lied
.
T
h
e
d
ataset
is
t
h
e
n
s
p
lit
i
n
to
tr
ain
in
g
an
d
test
i
n
g
s
u
b
s
ets
(
8
0
%
-
2
0
%),
en
s
u
r
in
g
ef
f
ec
ti
v
e
ev
alu
atio
n
o
f
th
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
o
n
u
n
s
ee
n
d
ata.
T
h
ese
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
p
r
o
v
id
e
a
r
o
b
u
s
t
an
d
s
t
an
d
ar
d
ized
in
p
u
t
f
o
r
th
e
cG
A
N,
e
n
ab
lin
g
ac
c
u
r
ate
tr
an
s
l
atio
n
o
f
h
an
d
-
d
r
a
w
n
s
k
etc
h
es i
n
to
r
ea
lis
tic
i
m
a
g
es
[
2
5
]
,
[
2
6
]
.
3
.
2
.
E
v
a
lua
t
i
o
n
m
et
rics a
nd
re
s
ults
T
h
e
FID
s
co
r
e
m
ea
s
u
r
e
s
th
e
r
ese
m
b
la
n
ce
b
et
w
ee
n
t
h
e
d
is
tr
i
b
u
tio
n
o
f
t
h
e
g
en
er
ated
i
m
ag
e
s
an
d
th
e
r
ea
l
im
a
g
e
s
,
w
it
h
lo
w
er
FID
v
alu
e
s
in
d
icati
n
g
g
r
ea
ter
r
ea
lis
m
.
I
n
t
h
is
s
t
u
d
y
,
t
h
e
FID
s
co
r
e
is
tr
ac
k
ed
ac
r
o
s
s
tr
ain
i
n
g
ep
o
ch
s
to
ev
al
u
ate
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
cG
AN
m
o
d
el.
Sh
o
w
n
h
er
e
i
n
Fi
g
u
r
e
5
th
e
FID
s
co
r
e
d
ec
r
ea
s
es
s
tead
il
y
as
th
e
tr
ai
n
in
g
p
r
o
g
r
ess
e
s
,
s
i
g
n
i
f
y
i
n
g
t
h
at
th
e
g
en
er
ato
r
n
et
w
o
r
k
is
i
m
p
r
o
v
in
g
its
ab
ilit
y
to
s
y
n
t
h
esize
r
ea
li
s
tic
h
u
m
a
n
f
ac
es f
r
o
m
h
a
n
d
-
d
r
a
w
n
s
k
etc
h
es.
T
h
is
s
tead
y
d
ec
lin
e
h
i
g
h
lig
h
t
s
th
e
m
o
d
el
’
s
ab
ilit
y
to
lear
n
an
d
ad
ap
t,
r
ed
u
cin
g
t
h
e
d
is
tan
ce
b
et
w
ee
n
t
h
e
g
en
er
at
ed
an
d
r
ea
l im
a
g
e
d
is
tr
ib
u
tio
n
s
.
Fig
u
r
e
5
.
FID
s
co
r
e
v
s
e
p
o
ch
s
I
n
o
u
r
r
esear
ch
,
th
e
L
1
lo
s
s
is
tr
ac
k
ed
to
ev
alu
ate
h
o
w
e
f
f
ec
t
iv
el
y
t
h
e
g
e
n
e
r
ato
r
n
et
w
o
r
k
r
ec
o
n
s
tr
u
ct
s
r
ea
lis
tic
i
m
ag
e
s
f
r
o
m
h
a
n
d
-
d
r
a
w
n
s
k
etc
h
es.
A
s
ill
u
s
tr
ated
in
Fi
g
u
r
e
6
th
e
L
1
lo
s
s
d
ec
r
ea
s
es
s
tead
il
y
o
v
er
th
e
tr
ain
i
n
g
ep
o
ch
s
,
in
d
icat
in
g
th
at
th
e
g
en
er
ato
r
is
i
m
p
r
o
v
i
n
g
its
ab
ilit
y
to
p
r
o
d
u
ce
h
ig
h
-
f
i
d
elit
y
i
m
a
g
es
t
h
at
clo
s
el
y
r
e
s
e
m
b
le
t
h
e
tar
g
et
o
u
tp
u
t
s
.
I
n
o
u
r
ex
p
er
i
m
en
t,
th
e
d
ec
lin
in
g
L
1
lo
s
s
d
e
m
o
n
s
tr
ates
th
e
g
e
n
er
ato
r
’
s
p
r
o
g
r
ess
iv
e
lear
n
in
g
an
d
ad
ap
tatio
n
to
th
e
s
k
e
tch
-
to
-
i
m
a
g
e
s
y
n
th
e
s
is
ta
s
k
.
E
v
e
n
t
u
all
y
,
th
e
lo
s
s
s
tab
ilizes
,
s
ig
n
i
f
y
in
g
t
h
at
th
e
g
e
n
er
ato
r
h
as r
ea
ch
ed
a
lev
el
o
f
co
n
s
is
ten
t p
er
f
o
r
m
a
n
ce
i
n
g
e
n
er
ati
n
g
r
e
alis
tic
i
m
a
g
es.
T
h
e
ad
v
er
s
ar
ial
lo
s
s
is
m
o
n
ito
r
ed
to
ass
ess
t
h
e
d
is
cr
i
m
i
n
ato
r
’
s
p
er
f
o
r
m
a
n
ce
in
d
i
s
ti
n
g
u
i
s
h
i
n
g
b
et
w
ee
n
r
ea
l
an
d
g
en
er
ated
i
m
ag
e
s
.
As
illu
s
tr
ated
in
Fi
g
u
r
e
7
th
e
ad
v
er
s
ar
ial
lo
s
s
d
ec
r
ea
s
es
o
v
er
th
e
tr
ain
i
n
g
ep
o
ch
s
,
r
ef
lecti
n
g
t
h
e
d
is
cr
i
m
i
n
ato
r
’
s
ab
ilit
y
to
e
f
f
ec
t
iv
el
y
d
if
f
er
en
t
iate
b
et
w
ee
n
r
ea
l
an
d
s
y
n
t
h
es
ized
im
a
g
es.
T
h
is
b
eh
av
io
r
in
d
icate
s
t
h
at
th
e
cG
A
N
i
s
tr
ain
i
n
g
s
u
cc
es
s
f
u
l
l
y
,
with
th
e
g
en
er
ato
r
i
m
p
r
o
v
i
n
g
it
s
o
u
tp
u
t
to
th
e
p
o
in
t
w
h
er
e
th
e
d
is
cr
i
m
i
n
ato
r
f
i
n
d
s
i
t in
cr
ea
s
i
n
g
l
y
c
h
alle
n
g
in
g
to
d
is
ce
r
n
g
e
n
er
ate
d
p
h
o
to
s
f
r
o
m
r
ea
l o
n
es.
I
n
o
u
r
s
t
u
d
y
,
t
h
e
r
ea
li
s
m
s
co
r
e
is
tr
ac
k
ed
to
ev
a
lu
ate
th
e
p
r
o
g
r
ess
io
n
o
f
t
h
e
g
e
n
er
ated
i
m
ag
e
s
’
q
u
ali
t
y
o
v
er
th
e
tr
ain
in
g
p
r
o
ce
s
s
.
A
s
s
h
o
wn
in
Fig
u
r
e
8
th
e
r
ea
lis
m
s
co
r
e
i
m
p
r
o
v
es st
ea
d
il
y
ac
r
o
s
s
ep
o
ch
s
,
h
ig
h
li
g
h
ti
n
g
t
h
e
g
en
er
ato
r
’
s
in
cr
ea
s
in
g
ab
ilit
y
to
p
r
o
d
u
ce
h
ig
h
-
f
id
eli
t
y
i
m
a
g
es
t
h
at
clo
s
el
y
r
e
s
e
m
b
le
t
h
e
r
ea
l
im
a
g
e
s
.
T
h
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
23
,
No
.
4
,
A
u
g
u
s
t
20
25
:
9
7
6
-
9
85
982
d
em
o
n
s
tr
ate
s
th
e
e
f
f
ec
tiv
e
n
es
s
o
f
th
e
cG
A
N
m
o
d
el
i
n
tr
an
s
f
o
r
m
in
g
h
a
n
d
-
d
r
a
w
n
s
k
e
tch
e
s
in
to
r
ea
lis
tic
a
n
d
d
etailed
h
u
m
a
n
f
ac
e
s
,
u
n
d
er
s
co
r
in
g
t
h
e
s
u
cc
es
s
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
.
T
ab
le
1
s
u
m
m
ar
izes
k
e
y
m
e
tr
ics,
in
cl
u
d
in
g
FID
,
L
1
lo
s
s
,
g
en
er
ato
r
lo
s
s
,
d
is
cr
i
m
i
n
ato
r
lo
s
s
,
a
n
d
r
ea
lis
m
s
co
r
e
,
ev
alu
ated
o
v
er
d
if
f
er
en
t
tr
ain
i
n
g
ep
o
ch
s
.
I
t
h
i
g
h
l
ig
h
ts
th
e
p
r
o
g
r
ess
i
v
e
i
m
p
r
o
v
e
m
e
n
t
i
n
t
h
e
q
u
a
lit
y
a
n
d
r
ea
lis
m
o
f
t
h
e
g
en
er
ate
d
i
m
ag
e
s
as
tr
ai
n
i
n
g
ad
v
an
ce
s
.
Fig
u
r
e
6
.
L
1
l
o
s
s
v
s
e
p
o
ch
s
Fig
u
r
e
7
.
A
d
v
er
s
ar
ia
l
l
o
s
s
v
s
e
p
o
ch
s
Fig
u
r
e
8
.
R
ea
lis
m
s
co
r
e
v
s
e
p
o
ch
s
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
E
n
h
a
n
cin
g
r
ea
lis
m
in
h
a
n
d
-
d
r
a
w
n
h
u
ma
n
s
ke
tch
es th
r
o
u
g
h
co
n
d
itio
n
a
l g
en
era
tive
…
(
I
mra
n
Ulla
K
h
a
n
)
983
T
ab
le
1
.
P
er
f
o
r
m
a
n
ce
m
etr
ic
s
ac
r
o
s
s
tr
ain
i
n
g
ep
o
ch
s
Ep
o
c
h
F
I
D
s
c
o
r
e
L
1
l
o
ss
G
e
n
e
r
a
t
o
r
l
o
ss
D
i
scri
m
i
n
a
t
o
r
l
o
ss
R
e
a
l
i
sm
s
c
o
r
e
1
5
0
.
0
0
1
.
0
0
2
.
0
0
0
.
8
0
4
.
0
0
2
4
7
.
8
9
0
.
9
6
1
.
9
4
0
.
8
6
4
.
2
4
3
4
5
.
7
9
0
.
9
2
1
.
8
7
0
.
9
3
4
.
4
7
4
4
3
.
6
8
0
.
8
7
1
.
8
1
0
.
9
9
4
.
7
1
5
4
1
.
5
8
0
.
8
3
1
.
7
5
1
.
0
5
4
.
9
5
10
3
1
.
0
5
0
.
6
2
1
.
4
3
1
.
3
7
6
.
1
3
11
2
8
.
9
5
0
.
5
8
1
.
3
7
1
.
4
3
6
.
3
7
12
2
6
.
8
4
0
.
5
4
1
.
3
1
1
.
4
9
6
.
6
1
13
2
4
.
7
4
0
.
4
9
1
.
2
4
1
.
5
6
6
.
8
4
14
2
2
.
6
3
0
.
4
5
1
.
1
8
1
.
6
2
7
.
0
8
15
2
0
.
5
3
0
.
4
1
1
.
1
2
1
.
6
8
7
.
3
2
16
1
8
.
4
2
0
.
3
7
1
.
0
5
1
.
7
5
7
.
5
5
17
1
6
.
3
2
0
.
3
3
0
.
9
9
1
.
8
1
7
.
7
9
18
1
4
.
2
1
0
.
2
8
0
.
9
3
1
.
8
7
8
.
0
3
19
1
2
.
1
1
0
.
2
4
0
.
8
6
1
.
9
4
8
.
2
6
20
1
0
.
0
0
0
.
2
0
0
.
8
0
2
.
0
0
8
.
5
0
4.
CO
NCLU
SI
O
N
T
h
is
r
esear
ch
d
e
m
o
n
s
tr
ates
th
e
ef
f
ec
ti
v
en
e
s
s
o
f
cG
ANs
in
b
o
o
s
tin
g
t
h
e
ac
u
r
ac
y
o
f
h
a
n
d
-
d
r
aw
n
h
u
m
a
n
s
k
etc
h
es
b
y
g
e
n
er
atin
g
h
ig
h
-
f
id
elit
y
i
m
ag
e
s
.
T
h
r
o
u
g
h
tr
ai
n
in
g
o
n
C
UF
S
d
ataset,
th
e
m
o
d
el
s
ig
n
i
f
ica
n
tl
y
i
m
p
r
o
v
es
s
tr
u
ct
u
r
al
ac
cu
r
ac
y
a
n
d
f
in
e
d
etail
s
,
as
r
ef
lecte
d
in
en
h
a
n
ce
d
FID
s
co
r
es
an
d
q
u
alitativ
e
as
s
ess
m
e
n
ts
.
T
h
ese
f
in
d
i
n
g
s
h
a
v
e
s
i
g
n
i
f
ican
t
i
m
p
licatio
n
s
f
o
r
f
o
r
en
s
ic
i
n
v
esti
g
a
tio
n
s
,
s
u
p
p
o
r
tin
g
la
w
e
n
f
o
r
ce
m
e
n
t
ag
e
n
cie
s
in
id
en
ti
f
y
in
g
s
u
s
p
ec
ts
,
as
w
e
ll
as
b
en
ef
itin
g
cr
ea
tiv
e
in
d
u
s
tr
i
es
th
r
o
u
g
h
au
to
m
ated
s
k
etc
h
-
to
-
i
m
a
g
e
tr
an
s
latio
n
.
Ho
w
e
v
er
,
li
m
itatio
n
s
s
u
ch
as
d
ataset
d
iv
er
s
it
y
an
d
m
o
d
el
g
en
er
aliza
tio
n
r
eq
u
ir
e
f
u
r
th
er
ex
p
lo
r
atio
n
.
Fu
t
u
r
e
w
o
r
k
w
i
ll
e
m
p
h
asi
s
o
n
ex
p
a
n
d
in
g
t
h
e
d
ataset,
f
in
e
-
t
u
n
i
n
g
t
h
e
m
o
d
el
f
o
r
d
if
f
er
e
n
t
s
k
etch
s
t
y
les,
an
d
in
te
g
r
ati
n
g
ad
d
itio
n
al
ev
al
u
atio
n
m
etr
ics.
T
h
e
s
tu
d
y
r
ei
n
f
o
r
ce
s
th
e
tr
an
s
f
o
r
m
ati
v
e
p
o
ten
tial
o
f
A
I
i
n
b
r
id
g
i
n
g
th
e
g
a
p
b
et
w
ee
n
ar
tis
t
ic
s
k
etch
e
s
a
n
d
p
h
o
to
r
ea
lis
tic
i
m
a
g
er
y
,
r
ev
o
l
u
ti
o
n
izin
g
ap
p
licatio
n
s
i
n
d
ig
i
tal
ar
tis
tr
y
a
n
d
f
o
r
en
s
ic
tech
n
o
lo
g
y
.
ACK
NO
WL
E
D
G
M
E
NT
S
T
h
e
au
th
o
r
s
ar
e
g
r
ate
f
u
l
to
R
E
VA
U
n
i
v
er
s
it
y
f
o
r
p
r
o
v
id
in
g
t
h
e
n
ec
e
s
s
ar
y
r
e
s
o
u
r
ce
s
a
n
d
f
a
cilities
f
o
r
th
is
s
t
u
d
y
.
F
UNDIN
G
I
NF
O
RM
AT
I
O
N
W
e
th
e
au
t
h
o
r
s
d
ec
lar
e
th
at
n
o
f
u
n
d
in
g
w
a
s
r
ec
eiv
ed
f
o
r
th
i
s
r
esear
ch
w
o
r
k
.
AUTHO
R
CO
NT
RIB
UT
I
O
NS ST
A
T
E
M
E
NT
T
h
is
j
o
u
r
n
al
u
s
e
s
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
i
v
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
t
h
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
lla
b
o
r
atio
n
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
I
m
r
an
Ulla
K
h
a
n
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Dep
a
R
a
m
ac
h
an
d
r
aia
h
Ku
m
ar
R
aj
a
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
si
s
I
:
I
n
v
e
st
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
si
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
t
h
o
r
s
s
tate
n
o
co
n
f
lic
t o
f
i
n
t
er
est.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
23
,
No
.
4
,
A
u
g
u
s
t
20
25
:
9
7
6
-
9
85
984
I
NF
O
RM
E
D
CO
NSE
N
T
W
e
h
av
e
o
b
tain
ed
in
f
o
r
m
ed
c
o
n
s
en
t f
r
o
m
al
l
in
d
i
v
id
u
al
s
in
c
lu
d
ed
in
th
is
s
t
u
d
y
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
i
n
g
s
o
f
th
is
s
t
u
d
y
ar
e
p
u
b
licl
y
a
v
ailab
le
f
r
o
m
t
h
e
C
UHK
Mu
lti
m
ed
ia
L
a
b
at
h
ttp
://
m
m
lab
.
ie.
cu
h
k
.
ed
u
.
h
k
/ar
ch
i
v
e/
f
ac
es
k
etc
h
.
h
t
m
l.
R
ef
er
en
ce
:
Z
h
a
n
g
,
W
an
g
,
a
n
d
T
an
g
,
513
-
5
2
0
.
1
0
.
1
1
0
9
/C
VP
R
.
2
0
1
1
.
5
9
9
5
3
2
4
RE
F
E
R
E
NC
E
S
[
1
]
I
.
U
.
K
h
a
n
a
n
d
R
.
D
.
R
.
K
u
mar,
“
R
e
v
i
e
w
o
n
r
e
a
l
t
i
me
a
p
p
r
o
a
c
h
t
o
i
d
e
n
t
i
f
y
a
p
e
r
so
n
b
a
se
d
o
n
h
a
n
d
d
r
a
w
n
s
k
e
t
c
h
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
i
n
2
0
2
3
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
v
e
n
t
i
v
e
R
e
se
a
rc
h
i
n
C
o
m
p
u
t
i
n
g
A
p
p
l
i
c
a
t
i
o
n
s (I
C
I
R
C
A)
,
A
u
g
.
2
0
2
3
,
p
p
.
4
9
3
–
4
9
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
R
C
A
5
7
9
8
0
.
2
0
2
3
.
1
0
2
2
0
7
0
4
.
[
2
]
H
.
Y
a
n
e
t
a
l
.
,
“
T
o
w
a
r
d
i
n
t
e
l
l
i
g
e
n
t
d
e
si
g
n
:
A
n
A
I
-
b
a
se
d
f
a
sh
i
o
n
d
e
si
g
n
e
r
u
si
n
g
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s
a
i
d
e
d
b
y
sk
e
t
c
h
a
n
d
r
e
n
d
e
r
i
n
g
g
e
n
e
r
a
t
o
r
s,
”
I
E
EE
T
r
a
n
sa
c
t
i
o
n
s
o
n
M
u
l
t
i
m
e
d
i
a
,
v
o
l
.
2
5
,
p
p
.
2
3
2
3
–
2
3
3
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
M
M
.
2
0
2
2
.
3
1
4
6
0
1
0
.
[
3
]
A
.
A
k
r
a
m,
N
.
W
a
n
g
,
X
.
G
a
o
,
a
n
d
J
.
L
i
,
“
I
n
t
e
g
r
a
t
i
n
g
G
A
N
w
i
t
h
C
N
N
f
o
r
f
a
c
e
sk
e
t
c
h
sy
n
t
h
e
si
s,
”
i
n
2
0
1
8
I
EE
E
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s (I
C
C
C
)
,
D
e
c
.
2
0
1
8
,
p
p
.
1
4
8
3
–
1
4
8
7
.
d
o
i
:
1
0
.
1
1
0
9
/
C
o
mp
C
o
mm
.
2
0
1
8
.
8
7
8
0
6
4
8
.
[
4
]
S
.
N
.
B
u
s
h
r
a
a
n
d
K
.
U
.
M
a
h
e
sw
a
r
i
,
“
C
r
i
me
i
n
v
e
st
i
g
a
t
i
o
n
u
si
n
g
D
C
G
A
N
b
y
f
o
r
e
n
si
c
s
k
e
t
c
h
-
to
-
f
a
c
e
t
r
a
n
sf
o
r
mat
i
o
n
(
S
T
F
)
-
A
r
e
v
i
e
w
,
”
i
n
2
0
2
1
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
i
n
g
Me
t
h
o
d
o
l
o
g
i
e
s a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
(
I
C
C
M
C
)
,
A
p
r
.
2
0
2
1
,
p
p
.
1
3
4
3
–
1
3
4
8
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
M
C
5
1
0
1
9
.
2
0
2
1
.
9
4
1
8
4
1
7
.
[
5
]
K
.
S
.
K
i
t
,
W
.
K
.
W
o
n
g
,
I
.
M
.
C
h
e
w
,
F
.
H
.
Ju
w
o
n
o
,
a
n
d
S
.
S
i
v
a
k
u
mar
,
“
A
sc
o
p
i
n
g
r
e
v
i
e
w
o
f
G
A
N
-
g
e
n
e
r
a
t
e
d
i
mag
e
s d
e
t
e
c
t
i
o
n
,
”
i
n
2
0
2
3
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
D
i
g
i
t
a
l
A
p
p
l
i
c
a
t
i
o
n
s,
T
r
a
n
s
f
o
rm
a
t
i
o
n
&
Ec
o
n
o
m
y
(
I
C
D
ATE)
,
J
u
l
.
2
0
2
3
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
D
A
T
E5
8
1
4
6
.
2
0
2
3
.
1
0
2
4
8
6
7
9
.
[
6
]
A
.
P
o
d
d
a
r
,
S
.
G
a
w
a
d
e
,
P
.
V
a
r
p
e
,
a
n
d
S
.
B
h
a
g
w
a
t
,
“
F
r
o
n
t
a
l
f
a
c
e
l
a
n
d
mar
k
g
e
n
e
r
a
t
i
o
n
u
s
i
n
g
G
A
N
,
”
i
n
2
0
2
2
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
p
p
l
i
e
d
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
C
o
m
p
u
t
i
n
g
(
I
C
AAIC)
,
M
a
y
2
0
2
2
,
p
p
.
1
1
7
2
–
1
1
7
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
A
I
C
5
3
9
2
9
.
2
0
2
2
.
9
7
9
3
1
8
9
.
[
7
]
B
.
K
u
r
i
a
k
o
se
,
T
.
T
h
o
mas,
N
.
E.
T
h
o
m
a
s,
S
.
J.
V
a
r
g
h
e
se
,
a
n
d
V
.
A
.
K
u
m
a
r
,
“
S
y
n
t
h
e
si
z
i
n
g
i
mag
e
s fr
o
m h
a
n
d
-
d
r
a
w
n
sk
e
t
c
h
e
s u
s
i
n
g
c
o
n
d
i
t
i
o
n
a
l
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s,
”
i
n
2
0
2
0
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
E
l
e
c
t
r
o
n
i
c
s
a
n
d
S
u
s
t
a
i
n
a
b
l
e
C
o
m
m
u
n
i
c
a
t
i
o
n
S
y
s
t
e
m
s (I
C
ES
C
)
,
J
u
l
.
2
0
2
0
,
p
p
.
7
7
4
–
7
7
8
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
ESC4
8
9
1
5
.
2
0
2
0
.
9
1
5
5
5
5
0.
[
8
]
R
.
B
a
y
o
u
mi
,
M
.
A
l
f
o
n
se
,
a
n
d
A
.
-
B
.
M
.
S
a
l
e
m,
“
A
n
i
n
t
e
l
l
i
g
e
n
t
h
y
b
r
i
d
t
e
x
t
-
to
-
i
mag
e
sy
n
t
h
e
si
s
mo
d
e
l
f
o
r
g
e
n
e
r
a
t
i
n
g
r
e
a
l
i
st
i
c
h
u
m
a
n
f
a
c
e
s,
”
i
n
2
0
2
1
T
e
n
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
(
I
C
I
C
I
S
)
,
D
e
c
.
2
0
2
1
,
p
p
.
1
7
2
–
1
7
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
C
I
S
5
2
5
9
2
.
2
0
2
1
.
9
6
9
4
1
9
4
.
[
9
]
M
.
V
i
j
a
y
,
M
.
M
e
g
h
a
n
a
,
N
.
A
k
l
e
c
h
a
,
a
n
d
R
.
S
r
i
n
a
t
h
,
“
D
i
a
l
o
g
d
r
i
v
e
n
f
a
c
e
c
o
n
s
t
r
u
c
t
i
o
n
u
s
i
n
g
G
A
N
s,
”
i
n
2
0
2
0
I
EE
E
3
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
T
o
o
l
s
w
i
t
h
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(
I
C
T
AI
)
,
N
o
v
.
2
0
2
0
,
p
p
.
6
4
7
–
6
5
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
TA
I
5
0
0
4
0
.
2
0
2
0
.
0
0
1
0
4
.
[
1
0
]
J.
L
i
,
T
.
S
u
n
,
Z
.
Y
a
n
g
,
a
n
d
Z
.
Y
u
a
n
,
“
M
e
t
h
o
d
s a
n
d
d
a
t
a
se
t
s
o
f
t
e
x
t
t
o
i
m
a
g
e
sy
n
t
h
e
si
s
b
a
se
d
o
n
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
2
I
E
EE
5
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
a
n
d
C
o
m
p
u
t
e
r
A
i
d
e
d
E
d
u
c
a
t
i
o
n
(
I
C
I
S
C
AE)
,
S
e
p
.
2
0
2
2
,
p
p
.
8
4
3
–
8
4
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
S
C
A
E5
5
8
9
1
.
2
0
2
2
.
9
9
2
7
6
3
4
.
[
1
1
]
K
.
V
.
S
w
a
my
,
A
.
S
u
p
r
a
j
a
,
P
.
S
.
V
i
n
u
t
h
n
a
,
a
n
d
D
.
L
.
S
i
n
d
h
u
r
a
,
“
P
e
r
f
o
r
m
a
n
c
e
c
o
m
p
a
r
i
so
n
o
f
v
a
r
i
o
u
s
f
e
a
t
u
r
e
s
f
o
r
h
u
ma
n
f
a
c
e
r
e
c
o
g
n
i
t
i
o
n
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
in
2
0
2
2
I
EE
E
C
o
n
f
e
re
n
c
e
o
n
I
n
t
e
r
d
i
s
c
i
p
l
i
n
a
ry
Ap
p
r
o
a
c
h
e
s
i
n
T
e
c
h
n
o
l
o
g
y
a
n
d
Ma
n
a
g
e
m
e
n
t
f
o
r
S
o
c
i
a
l
I
n
n
o
v
a
t
i
o
n
(
I
ATM
S
I
)
,
D
e
c
.
2
0
2
2
,
p
p
.
1
–
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
A
T
M
S
I
5
6
4
5
5
.
2
0
2
2
.
1
0
1
1
9
4
4
9
.
[
1
2
]
V
.
s
.
K
.
K
a
t
t
a
,
H
.
K
a
p
a
l
a
v
a
i
,
a
n
d
S
.
M
o
n
d
a
l
,
“
G
e
n
e
r
a
t
i
n
g
n
e
w
h
u
man
f
a
c
e
s
a
n
d
i
mp
r
o
v
i
n
g
t
h
e
q
u
a
l
i
t
y
o
f
i
mag
e
s
u
si
n
g
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s
(
G
A
N
)
,
”
i
n
2
0
2
3
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
E
d
g
e
C
o
m
p
u
t
i
n
g
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
(
I
C
EC
AA)
,
J
u
l
.
2
0
2
3
,
p
p
.
1
6
4
7
–
1
6
5
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EC
A
A
5
8
1
0
4
.
2
0
2
3
.
1
0
2
1
2
0
9
9
.
[
1
3
]
R
.
J
a
d
h
a
v
,
V
.
G
o
k
h
a
l
e
,
M
.
D
e
sh
p
a
n
d
e
,
A
.
G
o
r
e
,
A
.
G
h
a
r
p
u
r
e
,
a
n
d
H
.
Y
a
d
a
v
,
“
H
i
g
h
f
i
d
e
l
i
t
y
f
a
c
e
g
e
n
e
r
a
t
i
o
n
w
i
t
h
st
y
l
e
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s,
”
i
n
2
0
2
3
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
m
a
r
t
T
e
c
h
n
o
l
o
g
i
e
s a
n
d
S
y
st
e
m
s
f
o
r N
e
x
t
G
e
n
e
ra
t
i
o
n
C
o
m
p
u
t
i
n
g
(
I
C
S
T
S
N
)
,
A
p
r
.
2
0
2
3
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
T
S
N
5
7
8
7
3
.
2
0
2
3
.
1
0
1
5
1
6
0
3
.
[
1
4
]
K
.
K
o
v
a
r
t
h
a
n
a
n
a
n
d
K
.
M
.
S
.
J
.
K
u
marasi
n
g
h
e
,
“
G
e
n
e
r
a
t
i
n
g
p
h
o
t
o
g
r
a
p
h
i
c
f
a
c
e
i
mag
e
s
f
r
o
m
sk
e
t
c
h
e
s:
A
st
u
d
y
o
f
G
A
N
-
b
a
se
d
a
p
p
r
o
a
c
h
e
s,
”
i
n
2
0
2
3
8
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
Re
s
e
a
r
c
h
(
I
C
I
T
R)
,
D
e
c
.
2
0
2
3
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
T
R
6
1
0
6
2
.
2
0
2
3
.
1
0
3
8
2
9
4
4
.
[
1
5
]
S
.
S
,
R
.
P
.
K
u
mar,
a
n
d
S
.
N
.
M
u
d
a
ssi
r
,
“
S
k
e
t
c
h
t
o
i
mag
e
sy
n
t
h
e
si
s
u
s
i
n
g
a
t
t
e
n
t
i
o
n
b
a
se
d
c
o
n
t
e
x
t
u
a
l
G
A
N
,
”
i
n
2
0
2
3
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
i
n
g
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
N
e
t
w
o
r
k
i
n
g
T
e
c
h
n
o
l
o
g
i
e
s
(
I
C
C
C
N
T
)
,
J
u
l
.
2
0
2
3
,
p
p
.
1
–
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
N
T
5
6
9
9
8
.
2
0
2
3
.
1
0
3
0
6
4
4
4
.
[
1
6
]
A
.
K
.
B
h
u
n
i
a
e
t
a
l
.
,
“
S
k
e
t
c
h
2
S
a
l
i
e
n
c
y
:
L
e
a
r
n
i
n
g
t
o
d
e
t
e
c
t
s
a
l
i
e
n
t
o
b
j
e
c
t
s
f
r
o
m
h
u
m
a
n
d
r
a
w
i
n
g
s,
”
i
n
2
0
2
3
I
E
EE/
C
VF
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
a
n
d
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
(
C
VP
R)
,
J
u
n
.
2
0
2
3
,
p
p
.
2
7
3
3
–
2
7
4
3
.
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
5
2
7
2
9
.
2
0
2
3
.
0
0
2
6
8
.
[
1
7
]
M
.
A
.
K
h
a
n
a
n
d
A
.
S
.
Jal
a
l
,
“
S
u
s
p
e
c
t
i
d
e
n
t
i
f
i
c
a
t
i
o
n
u
s
i
n
g
l
o
c
a
l
f
a
c
i
a
l
a
t
t
r
i
b
u
t
e
d
b
y
f
u
si
n
g
f
a
c
i
a
l
l
a
n
d
m
a
r
k
s o
n
t
h
e
f
o
r
e
n
si
c
sk
e
t
c
h
,
”
i
n
2
0
2
0
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
n
t
e
m
p
o
r
a
ry
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
(
I
C
3
A)
,
F
e
b
.
2
0
2
0
,
p
p
.
1
8
1
–
1
8
6
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
3
A
4
8
9
5
8
.
2
0
2
0
.
2
3
3
2
9
3
.
[
1
8
]
C
.
P
e
n
g
,
C
.
Z
h
a
n
g
,
D
.
L
i
u
,
N
.
W
a
n
g
,
a
n
d
X
.
G
a
o
,
“
H
i
F
i
S
k
e
t
c
h
:
H
i
g
h
f
i
d
e
l
i
t
y
f
a
c
e
p
h
o
t
o
-
sk
e
t
c
h
sy
n
t
h
e
si
s
a
n
d
m
a
n
i
p
u
l
a
t
i
o
n
,
”
I
E
E
E
T
ra
n
s
a
c
t
i
o
n
s
o
n
I
m
a
g
e
Pro
c
e
ss
i
n
g
, v
o
l
.
3
2
,
p
p
.
5
8
6
5
–
5
8
7
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
T
I
P
.
2
0
2
3
.
3
3
2
6
6
8
0
.
[
1
9
]
J.
Z
h
e
n
g
,
Y
.
T
a
n
g
,
A
.
H
u
a
n
g
,
a
n
d
D
.
W
u
,
“
H
i
e
r
a
r
c
h
i
c
a
l
mu
l
t
i
v
a
r
i
a
t
e
r
e
p
r
e
se
n
t
a
t
i
o
n
l
e
a
r
n
i
n
g
f
o
r
f
a
c
e
sk
e
t
c
h
r
e
c
o
g
n
i
t
i
o
n
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
Em
e
r
g
i
n
g
T
o
p
i
c
s
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
8
,
n
o
.
2
,
p
p
.
2
0
3
7
–
2
0
4
9
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
T
ET
C
I
.
2
0
2
4
.
3
3
5
9
0
9
0
.
[
2
0
]
W
.
F
e
n
g
,
Z
.
M
e
n
g
,
a
n
d
L
.
W
a
n
g
,
“
S
t
a
c
k
e
d
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s
f
o
r
i
mag
e
g
e
n
e
r
a
t
i
o
n
b
a
se
d
o
n
U
-
N
e
t
d
i
s
c
r
i
mi
n
a
t
o
r
,
”
i
n
2
0
2
2
Asi
a
C
o
n
f
e
re
n
c
e
o
n
A
l
g
o
ri
t
h
m
s,
C
o
m
p
u
t
i
n
g
a
n
d
Ma
c
h
i
n
e
L
e
a
rn
i
n
g
(
C
AC
ML)
,
M
a
r
.
2
0
2
2
,
p
p
.
7
6
2
–
7
6
8
.
d
o
i
:
1
0
.
1
1
0
9
/
C
A
C
M
L
5
5
0
7
4
.
2
0
2
2
.
0
0
1
3
2
.
[
2
1
]
S
.
Z
h
a
n
g
,
H
.
W
a
n
g
,
a
n
d
L
.
W
a
n
g
,
“
A
s
e
n
si
t
i
v
e
i
mag
e
g
e
n
e
r
a
t
i
o
n
me
t
h
o
d
b
a
se
d
o
n
i
m
p
r
o
v
e
d
P
a
t
c
h
G
A
N
,
”
i
n
2
0
2
3
1
2
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
f
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
(
I
C
T
e
c
h
)
,
A
p
r
.
2
0
2
3
,
p
p
.
5
6
8
–
5
7
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
T
e
c
h
5
8
3
6
2
.
2
0
2
3
.
0
0
1
1
0
.
[
2
2
]
A
.
S
.
S
a
q
l
a
i
n
,
F
.
F
a
n
g
,
T
.
A
h
mad
,
L
.
W
a
n
g
,
a
n
d
Z
.
A
b
i
d
i
n
,
“
Ev
o
l
u
t
i
o
n
a
n
d
e
f
f
e
c
t
i
v
e
n
e
ss
o
f
l
o
ss
f
u
n
c
t
i
o
n
s
i
n
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s,
”
C
h
i
n
a
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
8
,
n
o
.
1
0
,
p
p
.
4
5
–
7
6
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
2
3
9
1
9
/
J
C
C
.
2
0
2
1
.
1
0
.
0
0
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
E
n
h
a
n
cin
g
r
ea
lis
m
in
h
a
n
d
-
d
r
a
w
n
h
u
ma
n
s
ke
tch
es th
r
o
u
g
h
co
n
d
itio
n
a
l g
en
era
tive
…
(
I
mra
n
Ulla
K
h
a
n
)
985
[
2
3
]
B
.
X
i
e
a
n
d
C
.
Ju
n
g
,
“
D
e
e
p
f
a
c
e
g
e
n
e
r
a
t
i
o
n
f
r
o
m
a
r
o
u
g
h
s
k
e
t
c
h
u
si
n
g
mu
l
t
i
-
l
e
v
e
l
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s,
”
i
n
2
0
2
2
2
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
P
a
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
(
I
C
PR)
, A
u
g
.
2
0
2
2
,
p
p
.
1
2
0
0
–
1
2
0
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
R
5
6
3
6
1
.
2
0
2
2
.
9
9
5
6
1
2
6
.
[
2
4
]
G
.
S
r
u
j
a
n
a
,
Y
.
M
a
d
h
u
r
i
,
T
.
H
a
r
s
h
i
t
h
a
,
U
.
G
.
M
a
n
o
h
a
r
i
,
a
n
d
S
.
R
a
n
i
,
“
S
u
s
p
e
c
t
f
a
c
e
d
e
t
e
c
t
i
o
n
b
y
a
u
t
o
sk
e
t
c
h
i
n
g
,
”
i
n
2
0
2
3
I
EEE
1
2
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
S
y
st
e
m
s
a
n
d
N
e
t
w
o
r
k
T
e
c
h
n
o
l
o
g
i
e
s
(
C
S
N
T
)
,
A
p
r
.
2
0
2
3
,
p
p
.
4
5
6
–
4
6
3
.
d
o
i
:
1
0
.
1
1
0
9
/
C
S
N
T
5
7
1
2
6
.
2
0
2
3
.
1
0
1
3
4
7
3
7
.
[
2
5
]
C
.
P
h
i
l
i
p
a
n
d
L
.
H
.
Jo
n
g
,
“
F
a
c
e
sk
e
t
c
h
sy
n
t
h
e
si
s
u
si
n
g
c
o
n
d
i
t
i
o
n
a
l
a
d
v
e
r
s
a
r
i
a
l
n
e
t
w
o
r
k
s,
”
i
n
2
0
1
7
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
C
o
n
v
e
r
g
e
n
c
e
(
I
C
T
C
)
,
O
c
t
.
2
0
1
7
,
p
p
.
3
7
3
–
3
7
8
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
T
C
.
2
0
1
7
.
8
1
9
1
0
0
6
.
[
2
6
]
Z
.
L
i
,
C
.
D
e
n
g
,
E.
Y
a
n
g
,
a
n
d
D
.
T
a
o
,
“
S
t
a
g
e
d
sk
e
t
c
h
-
to
-
i
m
a
g
e
s
y
n
t
h
e
si
s
v
i
a
se
mi
-
su
p
e
r
v
i
se
d
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
sari
a
l
n
e
t
w
o
r
k
s,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
M
u
l
t
i
m
e
d
i
a
,
v
o
l
.
2
3
,
p
p
.
2
6
9
4
–
2
7
0
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
T
M
M
.
2
0
2
0
.
3
0
1
5
0
1
5
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
I
m
r
a
n
Ulla
K
h
a
n
is
a
re
se
a
rc
h
s
c
h
o
lar
a
t
Re
v
a
Un
iv
e
rsit
y
a
n
d
a
n
A
ss
istan
t
P
r
o
f
e
ss
o
r
in
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
a
t
S
ri
Krish
n
a
In
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
,
Ba
n
g
a
lo
re
,
In
d
ia.
He
is
a
tec
h
e
n
t
h
u
sia
st
w
it
h
a
k
e
e
n
i
n
t
e
re
st
in
im
a
g
e
p
ro
c
e
ss
in
g
a
n
d
d
e
e
p
l
e
a
rn
i
n
g
.
His
re
se
a
rc
h
f
o
c
u
se
s
o
n
sk
e
tch
-
b
a
se
d
im
a
g
e
re
tri
e
v
a
l
a
n
d
g
e
n
e
ra
ti
v
e
a
d
v
e
rsa
rial
n
e
tw
o
rk
s
(GA
Ns
)
f
o
r
e
n
h
a
n
c
in
g
sk
e
tch
-
to
-
re
a
li
stic
im
a
g
e
tran
sf
o
r
m
a
ti
o
n
.
He
h
a
s
p
re
se
n
ted
a
n
d
p
u
b
li
sh
e
d
h
is
w
o
rk
in
v
a
rio
u
s
c
o
n
f
e
re
n
c
e
s
a
n
d
jo
u
rn
a
ls.
He
is
p
a
ss
io
n
a
te
a
b
o
u
t
a
d
v
a
n
c
in
g
h
is
e
x
p
e
rti
se
in
d
e
e
p
lea
rn
in
g
a
rc
h
it
e
c
tu
re
s
to
a
d
d
re
ss
c
h
a
ll
e
n
g
e
s
in
im
a
g
e
s
y
n
th
e
sis a
n
d
re
late
d
d
o
m
a
in
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
im
r
a
n
1
6
1
9
8
4
@g
m
a
il
.
c
o
m
.
De
p
a
Ra
m
a
c
h
a
n
d
r
a
ia
h
K
u
m
a
r
Ra
ja
is
c
u
rre
n
tl
y
w
o
rk
in
g
a
s
P
ro
f
e
ss
o
r
in
t
h
e
S
c
h
o
o
l
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
a
n
d
En
g
in
e
e
rin
g
a
t
RE
V
A
Un
iv
e
rsity
Ba
n
g
a
lu
ru
,
Ka
rn
a
tak
a
,
In
d
ia.
He
re
c
e
iv
e
d
h
is
Ba
c
h
e
lo
r
o
f
T
e
c
h
n
o
lo
g
y
(B.
T
e
c
h
)
f
ro
m
J
NTU
A
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
a
n
d
M
a
ste
r
o
f
T
e
c
h
n
o
l
o
g
y
(M
.
T
e
c
h
)
f
ro
m
Na
ti
o
n
a
l
In
stit
u
te
o
f
Tec
h
n
o
lo
g
y
Ka
rn
a
tak
a
(NI
T
K)
S
u
ra
th
k
a
l,
K
a
rn
a
tak
a
,
In
d
ia.
He
re
c
e
iv
e
d
a
Do
c
to
ra
te
o
f
P
h
il
o
s
o
p
h
y
(P
h
.
D.)
f
ro
m
S
t
P
e
ters
Un
iv
e
rsity
,
Ch
e
n
n
a
i,
In
d
ia
f
o
r
A
n
e
ffe
c
ti
v
e
c
o
n
tex
t
-
d
riv
e
n
re
c
o
m
m
e
n
d
e
r
s
y
ste
m
f
o
r
e
-
c
o
m
m
e
rc
e
a
p
p
li
c
a
ti
o
n
s.
He
d
id
P
o
st
Do
c
to
ra
l
re
se
a
rc
h
a
t
Un
iv
e
rsiti
T
e
k
n
ik
a
l
M
a
la
y
si
a
,
M
e
lak
a
,
M
a
la
y
sia
f
ro
m
S
e
p
tem
b
e
r
2
0
2
3
to
S
e
p
tem
b
e
r
2
0
2
4
.
He
re
c
e
i
v
e
d
f
u
n
d
in
g
a
m
o
u
n
t
o
f
1
8
0
0
0
USD
f
o
r
c
a
rry
in
g
o
u
t
re
se
a
r
c
h
p
ro
jec
ts o
n
e
f
o
r
1
6
0
0
0
USD
f
ro
m
REV
A
Un
iv
e
rsit
y
f
o
r
th
e
p
ro
jec
t
Hu
m
a
n
o
id
Ro
b
o
t
a
n
d
c
o
m
p
lete
d
th
e
p
ro
jec
t
in
2
0
2
2
a
n
d
a
n
o
th
e
r
f
o
r
2
0
0
0
USD
f
ro
m
S
re
e
V
id
y
a
n
ik
e
th
a
n
Ed
u
c
a
ti
o
n
a
l
T
ru
st
f
o
r
th
e
p
ro
jec
t
a
u
t
o
m
a
ti
o
n
o
f
a
re
a
to
rs
f
o
r
a
q
u
a
c
u
lt
u
re
u
sin
g
I
o
T
a
n
d
c
o
m
p
lete
d
th
e
p
ro
jec
t
in
2
0
1
9
.
Hi
s
re
se
a
rc
h
a
re
a
s
in
c
lu
d
e
t
h
e
i
n
tern
e
t
o
f
th
i
n
g
s,
d
a
ta
m
in
in
g
,
m
a
c
h
in
e
lea
rn
i
n
g
a
n
d
a
rti
f
icia
l
in
telli
g
e
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
u
m
a
rra
jad
r@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.