T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
6
,
Dec
em
b
er
2
0
2
1
,
p
p
.
18
57
~1
8
64
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i6
.
2
0
9
9
4
1857
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
H
uma
n activ
i
ty r
ecog
nition for sta
tic
a
nd dy
na
mic
activ
ity
using
conv
o
lutiona
l neural ne
twork
Ag
us
E
k
o
M
ina
rno
,
Wa
hy
u
Andh
y
k
a
K
us
um
a
,
Yo
g
a
An
g
g
i K
urnia
wa
n
Un
iv
e
rsitas
M
u
h
a
m
m
a
d
iy
a
h
M
a
lan
g
,
M
a
lan
g
,
In
d
o
n
e
sia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 1
5
,
2
0
2
0
R
ev
is
ed
J
u
n
2
8
,
2
0
2
1
Acc
ep
ted
J
u
l 9
,
2
0
2
1
Ev
a
lu
a
ted
a
c
ti
v
i
ty
a
s
a
d
e
tail
o
f
t
h
e
h
u
m
a
n
p
h
y
sic
a
l
m
o
v
e
m
e
n
t
h
a
s
b
e
c
o
m
e
a
lea
d
in
g
su
b
jec
t
f
o
r
re
se
a
rc
h
e
rs.
A
c
ti
v
it
y
re
c
o
g
n
it
i
o
n
a
p
p
li
c
a
ti
o
n
is
u
ti
li
z
e
d
in
se
v
e
ra
l
a
re
a
s,
su
c
h
a
s
li
v
in
g
,
h
e
a
l
th
,
g
a
m
e
,
m
e
d
ica
l,
re
h
a
b
i
li
tatio
n
,
a
n
d
o
th
e
r
sm
a
rt
h
o
m
e
sy
ste
m
a
p
p
li
c
a
ti
o
n
s
.
An
a
c
c
e
lero
m
e
t
e
r
wa
s
p
o
p
u
lar
se
n
so
rs
to
re
c
o
g
n
iz
e
t
h
e
a
c
ti
v
it
y
,
a
s
we
ll
a
s
a
g
y
ro
sc
o
p
e
,
wh
ic
h
c
a
n
b
e
e
m
b
e
d
d
e
d
i
n
a
sm
a
rtp
h
o
n
e
.
S
i
g
n
a
l
wa
s
g
e
n
e
ra
ted
fro
m
t
h
e
a
c
c
e
lero
m
e
ter
a
s
a
ti
m
e
-
se
ries
d
a
ta
is
a
n
a
c
tu
a
l
a
p
p
ro
a
c
h
li
k
e
a
h
u
m
a
n
a
c
ti
f
v
it
y
p
a
tt
e
r
n
.
M
o
ti
o
n
d
a
ta
h
a
v
e
a
c
q
u
ired
in
3
0
v
o
lu
n
tee
rs.
D
y
n
a
m
ic
a
c
ti
v
e
s
(wa
lk
i
n
g
,
wa
lk
i
n
g
u
p
sta
irs
,
wa
lk
i
n
g
d
o
wn
sta
irs
)
a
s
DA
a
n
d
sta
ti
c
a
c
t
iv
e
s
(lay
i
n
g
,
sta
n
d
i
n
g
,
sitt
i
n
g
)
a
s
S
A
we
re
c
o
ll
e
c
ted
fro
m
v
o
lu
n
tee
rs.
S
A
a
n
d
DA
it
'
s
a
c
h
a
ll
e
n
g
i
n
g
p
ro
b
le
m
with
th
e
d
iffere
n
t
si
g
n
a
l
p
a
tt
e
rn
s,
S
A
sig
n
a
ls
c
o
in
c
i
d
e
b
e
twe
e
n
a
c
ti
v
it
ies
b
u
t
wi
th
a
c
lea
r
th
re
sh
o
ld
,
o
th
e
rwise
th
e
DA
sig
n
a
l
is
c
lea
rly
d
istri
b
u
ted
b
u
t
with
a
n
a
d
jac
e
n
t
u
p
p
e
r
t
h
re
sh
o
l
d
.
T
h
e
p
ro
p
o
se
d
n
e
two
r
k
stru
c
t
u
re
a
c
h
iev
e
s
a
sig
n
ifi
c
a
n
t
p
e
rfo
rm
a
n
c
e
with
t
h
e
b
e
st
o
v
e
ra
ll
a
c
c
u
ra
c
y
o
f
9
7
%
.
Th
e
re
su
lt
in
d
ica
ted
th
e
a
b
il
it
y
o
f
t
h
e
m
o
d
e
l
fo
r
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
io
n
p
u
rp
o
se
s.
K
ey
w
o
r
d
s
:
Acc
eler
o
m
eter
C
NN
C
o
n
v
o
lu
tio
n
m
atr
ix
Gy
r
o
s
co
p
e
Hu
m
an
ac
tiv
ity
r
ec
o
g
n
itio
n
Hy
p
er
p
ar
a
m
eter
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
:
W
ah
y
u
An
d
h
y
k
a
Ku
s
u
m
a
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atics
Un
iv
er
s
itas
Mu
h
am
m
ad
iy
ah
Ma
lan
g
R
ay
a
T
lo
g
o
m
as St.
No
.
2
4
6
,
T
h
ir
d
C
am
p
u
s
Un
iv
er
s
itas
Mu
h
am
m
ad
iy
ah
Ma
lan
g
,
E
ast J
av
a
,
Ma
lan
g
,
6
5
1
4
4
,
I
n
d
o
n
esia
E
m
ail:
k
u
s
u
m
a.
wah
y
u
.
a@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
Hu
m
an
ac
tiv
ity
r
ec
o
g
n
itio
n
(
HAR)
i
s
th
e
u
s
e
o
f
k
n
o
wled
g
e
an
d
im
ag
e
m
o
d
els
f
o
r
m
o
d
el
in
g
ac
tiv
ity
an
d
s
en
s
o
r
d
ata
[
1
]
.
Hu
m
an
ac
tiv
ity
r
ec
o
g
n
itio
n
h
as
th
e
av
ailab
ilit
y
o
f
co
m
p
lete
s
en
s
o
r
s
s
o
th
at
th
ey
ca
n
r
ec
o
g
n
ize
h
u
m
an
ac
tiv
ities
s
u
ch
as si
t
tin
g
,
walk
in
g
,
s
leep
in
g
,
r
u
n
n
i
n
g
,
an
d
s
tan
d
in
g
.
HAR ca
n
also
b
e
u
s
ed
a
s
a
to
o
l to
d
iag
n
o
s
e
a
d
is
ea
s
e
[
2
]
,
ac
tiv
ity
r
ec
o
g
n
itio
n
[
3]
,
[
4
]
,
an
d
b
e
u
s
ed
in
th
e
m
ilit
ar
y
f
iel
d
[
5
]
.
A
p
io
n
ee
r
in
HAR
r
esear
ch
u
s
in
g
an
ac
ce
le
r
o
m
eter
was
p
u
b
lis
h
ed
in
th
e
9
0
s
[
6
]
.
Ho
wev
er
,
th
e
m
o
s
t
wid
ely
cited
r
esear
ch
was
ab
le
to
p
r
o
d
u
ce
s
atis
f
y
in
g
r
esu
lts
with
m
an
y
s
en
s
o
r
s
s
i
m
u
ltan
eo
u
s
ly
a
n
d
u
s
in
g
v
ar
io
u
s
alg
o
r
ith
m
s
ca
r
r
ied
o
u
t
b
y
B
ao
an
d
I
n
tile
[
7
]
.
C
lass
if
icatio
n
o
f
th
e
in
tr
o
d
u
ctio
n
o
f
h
u
m
a
n
ac
tiv
ities
u
s
in
g
s
en
s
o
r
s
th
at
v
ar
y
f
r
o
m
th
e
d
ev
ice
is
a
class
ic
p
r
o
b
lem
.
I
t
is
,
th
er
ef
o
r
e
,
im
p
o
r
tan
t
to
f
in
d
a
m
eth
o
d
f
o
r
th
e
p
r
o
p
e
r
r
e
co
g
n
itio
n
o
f
h
u
m
an
ac
tiv
ity
f
r
o
m
d
ev
ice
s
en
s
o
r
s
[
8
]
.
HAR
u
s
in
g
s
m
ar
tp
h
o
n
e
s
en
s
o
r
s
is
a
c
lass
ic
m
u
lti
-
v
ar
iate
ti
m
e
s
er
ies
cla
s
s
if
icatio
n
p
r
o
b
lem
,
wh
ich
u
tili
ze
s
1
D
s
en
s
o
r
s
ig
n
als
an
d
ex
tr
ac
ts
f
ea
tu
r
es
to
b
e
ab
le
to
r
ec
o
g
n
ize
ac
tiv
ities
b
y
u
tili
zin
g
class
if
icatio
n
.
Ver
y
litt
le
r
esear
ch
o
n
HAR
u
s
es
in
-
d
ep
th
lear
n
in
g
tech
n
iq
u
es
a
n
d
au
t
o
m
atic
f
ea
tu
r
e
ex
tr
ac
tio
n
m
ec
h
a
n
is
m
s
.
T
h
e
latest
b
r
ea
k
th
r
o
u
g
h
in
im
ag
e
an
d
s
o
u
n
d
r
ec
o
g
n
itio
n
h
as
r
esu
lted
in
a
n
ew
f
ield
o
f
r
esear
ch
th
at
attr
ac
ts
en
th
u
s
iast
ic
r
esear
ch
er
s
ca
lled
d
ee
p
lear
n
in
[
9
]
.
T
h
e
c
o
n
v
o
l
u
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN)
n
eu
r
al
n
etwo
r
k
,
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
6
,
Dec
em
b
e
r
2
0
2
1
:
18
57
-
18
64
1858
p
ar
ticu
lar
,
is
a
s
u
itab
le
alg
o
r
ith
m
f
o
r
im
a
g
e
an
d
s
o
u
n
d
r
ec
o
g
n
itio
n
.
B
u
t
n
o
t
o
n
ly
im
ag
es
an
d
s
o
u
n
d
s
ca
n
b
e
p
r
o
ce
s
s
ed
b
y
C
NN
but
th
e
H
AR
d
ataset
is
al
s
o
a
g
o
o
d
im
p
lem
en
tatio
n
wh
e
n
p
r
o
ce
s
s
in
g
it
u
s
in
g
tim
e
-
s
er
ies
d
ata
ty
p
es.
Pre
v
io
u
s
s
tu
d
ies
u
s
ed
HAR
f
r
o
m
v
ar
i
o
u
s
ty
p
es
o
f
s
en
s
o
r
s
ex
ce
p
t
f
o
r
ca
m
er
as
a
n
d
ac
ce
l
er
o
m
eter
s
,
g
y
r
o
s
co
p
e
s
en
s
o
r
s
u
s
in
g
elec
tr
o
m
y
o
g
r
ap
h
,
in
f
r
ar
ed
au
d
io
,
an
d
o
th
er
s
en
s
o
r
s
[
1
0
]
.
A
n
a
cc
ele
r
o
m
eter
h
as
s
ev
er
a
l
ad
v
an
tag
es,
lo
w
g
r
ass
,
ch
ea
p
e
r
.
W
ith
th
e
s
m
all
d
im
en
s
io
n
s
em
b
ed
d
e
d
in
a
s
m
ar
tp
h
o
n
e,
t
h
e
ac
ce
ler
o
m
eter
ca
n
ea
s
ily
m
ea
s
u
r
e
h
u
m
an
m
o
v
em
en
ts
.
I
t
c
an
b
e
u
s
ed
in
a
v
ar
iet
y
o
f
d
if
f
e
r
en
t
p
o
s
itio
n
s
s
u
ch
a
s
ar
m
s
,
waist,
h
ea
d
,
s
h
o
u
ld
er
s
,
p
o
ck
ets
[
1
1
]
.
Fu
en
tes
an
d
co
lleag
u
es
in
a
s
tu
d
y
en
titl
ed
"
o
n
lin
e
m
o
tio
n
r
ec
o
g
n
itio
n
u
s
in
g
an
ac
ce
ler
o
m
eter
in
a
m
o
b
ile
d
ev
ice"
u
s
es
n
eu
r
al
n
e
two
r
k
s
in
r
ec
o
g
n
itio
n
o
f
h
u
m
a
n
b
o
d
y
m
o
tio
n
[
1
2
]
,
wh
ile
k
h
a
n
u
s
es
th
e
d
ec
is
io
n
tr
ee
m
eth
o
d
in
r
ec
o
g
n
izi
n
g
h
u
m
an
b
o
d
y
m
o
v
em
en
ts
f
r
o
m
W
ii
r
em
o
te
d
ata
[
1
3
]
.
Ot
h
er
s
tu
d
ies
u
s
in
g
a
co
m
b
in
atio
n
o
f
C
NN
an
d
m
a
ch
in
e
lear
n
i
n
g
m
eth
o
d
s
ap
p
e
ar
in
T
ab
le
1
.
O
n
e
o
f
th
e
r
es
ea
r
ch
er
s
u
s
in
g
th
e
Un
iv
er
s
ity
o
f
C
alif
o
r
n
ia
I
r
v
in
e
(
UC
I
)
HAR
d
ataset
wh
ich
h
as
1
2
8
f
ea
t
u
r
es
is
R
o
n
ao
in
2
0
1
6
in
T
ab
le
2
in
h
is
s
tu
d
y
en
titl
ed
"
h
u
m
an
ac
tiv
ity
r
ec
o
g
n
itio
n
with
s
m
ar
tp
h
o
n
e
s
en
s
o
r
s
u
s
in
g
d
ee
p
lear
n
in
g
n
eu
r
al
n
etwo
r
k
s
"
p
r
o
d
u
ce
s
an
ac
cu
r
ac
y
v
alu
e
o
f
9
3
.
7
5
% u
s
in
g
th
e
C
NN
an
d
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
(
MPL
)
alg
o
r
ith
m
[
1
4
]
.
I
n
th
is
r
esear
c
h
wo
r
k
,
we
p
r
o
p
o
s
ed
co
n
v
o
l
u
tio
n
n
eu
r
al
n
et
wo
r
k
a
p
p
r
o
ac
h
es
f
o
r
h
u
m
an
b
o
d
y
m
o
tio
n
r
ec
o
g
n
itio
n
with
s
tatic
ac
tiv
ity
an
d
d
y
n
am
ic
d
atasets
.
T
h
e
m
ain
co
n
tr
ib
u
tio
n
o
f
th
is
p
ap
er
i
s
a
m
o
d
el
o
f
C
NN
ac
h
iev
ed
g
o
o
d
ac
cu
r
ac
y
co
m
p
ar
ed
to
p
r
ev
io
u
s
r
esear
ch
.
T
h
e
m
o
d
el
p
r
o
v
id
ed
v
ar
ia
b
le
p
ar
am
eter
s
th
at
m
atch
an
d
h
av
e
a
h
ig
h
ac
cu
r
ac
y
v
alu
e
f
o
r
th
e
d
ataset
d
y
n
am
ic
ac
ti
v
es
(
DA
)
a
n
d
s
tatic
ac
tiv
es
(
SA
)
.
C
o
m
b
i
n
a
t
i
o
n
o
f
D
A
a
n
d
S
A
m
o
d
e
l
s
t
o
cl
a
s
s
i
f
y
H
A
R
wi
t
h
6
cl
as
s
es
.
T
h
e
r
e
m
ai
n
i
n
g
s
e
ct
i
o
n
o
f
t
h
i
s
p
a
p
e
r
is
o
r
g
a
n
i
z
e
d
a
s
f
o
l
l
o
ws
:
m
a
t
e
r
i
a
ls
a
n
d
m
e
t
h
o
d
s
,
d
a
t
a
a
cq
u
i
s
i
ti
o
n
s
w
h
i
c
h
t
h
e
i
n
f
o
r
m
a
tio
n
a
b
o
u
t
p
r
o
p
o
s
e
d
m
e
t
h
o
d
s
a
r
e
g
i
v
e
n
i
n
s
ec
t
i
o
n
2
.
T
h
e
o
b
t
a
i
n
e
d
r
e
s
u
l
ts
a
n
d
d
is
cu
s
s
h
o
w
t
h
e
p
r
o
p
o
s
e
d
m
et
h
o
d
s
o
l
v
e
d
t
h
e
p
r
o
b
l
e
m
a
r
e
g
i
v
en
i
n
s
e
c
ti
o
n
3
.
T
h
e
c
o
n
c
l
u
s
i
o
n
a
b
o
u
t
p
r
o
b
l
e
m
r
e
s
u
lt
s
is
g
i
v
e
n
i
n
s
e
c
ti
o
n
4
.
T
ab
le
1
.
Pre
v
i
o
u
s
C
NN
r
esear
ch
No
M
e
t
hod
A
c
c
u
r
a
t
e
A
c
t
i
v
i
t
y
N
u
mb
e
r
o
f
F
e
a
t
u
r
e
s
1
D
e
c
i
s
i
o
n
Tr
e
e
(
D
T)
[
4
]
R
a
n
d
o
m
F
o
r
e
s
t
(
R
F
)
Ex
t
r
a
Tr
e
e
(
X
T)
K
-
N
e
a
r
e
st
N
e
i
g
h
b
o
r
(
K
N
N
)
Lo
g
i
s
t
i
c
R
e
g
r
e
si
o
n
(
LR
)
S
u
p
p
o
r
t
V
e
c
t
o
r
C
l
a
ssi
f
i
e
r
(
S
V
C
)
En
se
mb
l
e
V
o
t
e
C
l
a
ssi
f
i
e
r
9
3
.
4
4
%
9
6
.
7
3
%
9
6
.
6
8
%
9
6
.
2
1
%
9
8
.
4
0
%
9
3
.
8
6
%
9
7
.
6
0
%
La
y
i
n
g
,
S
t
a
n
d
i
n
g
,
S
i
t
t
i
n
g
,
W
a
l
k
i
n
g
,
W
a
l
k
i
n
g
U
p
st
a
i
r
s,
W
a
l
k
i
n
g
D
o
w
n
s
t
a
i
r
s
5
6
1
2
S
u
p
p
o
r
t
V
e
c
t
o
r
M
a
c
h
i
n
e
(
S
V
M
)
[
1
2
]
9
3
%
S
t
o
p
p
i
n
g
,
W
a
l
k
i
n
g
,
S
t
a
n
d
i
n
g
-
u
p
,
S
i
t
t
i
n
g
-
d
o
w
n
1
0
0
3
C
N
N
+
M
P
L
[
1
4
]
H
u
ma
n
C
r
a
f
t
e
d
F
e
a
t
u
r
e
s
(
H
C
F
)
+
A
r
t
i
f
i
c
i
a
l
N
e
u
r
a
l
N
e
t
w
o
r
k
(
ANN
)
H
C
F
+
S
V
M
9
3
.
7
5
%
8
2
.
2
7
%
7
7
.
6
6
%
W
a
l
k
i
n
g
,
U
p
sa
t
r
a
i
r
,
D
o
w
n
st
a
i
r
,
S
i
t
t
i
n
g
,
S
t
a
n
d
i
n
g
,
La
y
i
n
g
1
2
8
4
D
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
(
D
C
N
N
)
[
1
5
]
F
a
st
a
n
d
r
o
b
u
s
t
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
st
r
u
c
t
u
r
e
(
F
R
D
C
N
N
)
9
4
%
9
5
%
w
a
l
k
i
n
g
,
j
o
g
g
i
n
g
,
j
u
m
p
i
n
g
,
a
n
d
g
o
u
p
st
a
i
r
s
a
n
d
g
o
d
o
w
n
st
a
i
r
s,
si
t
t
i
n
g
,
st
a
n
d
i
n
g
,
l
a
y
i
n
g
t
o
t
h
e
l
e
f
t
a
n
d
r
i
g
h
t
si
d
e
,
a
n
d
l
a
y
i
n
g
s
u
p
i
n
e
a
n
d
p
r
o
n
e
2
4
8
5
C
N
N
+
LST
M
[
1
6
]
9
9
%
A
r
m w
a
v
e
,
H
a
mm
e
r
,
F
o
r
w
a
r
d
p
u
n
c
h
,
H
i
g
h
t
r
o
w
,
H
a
n
d
c
l
a
p
,
B
e
n
d
,
Te
n
n
i
s
serv
e
,
P
i
c
k
u
p
a
n
d
Tr
o
w
2
0
0
6
CN
N
[
1
7
]
C
o
n
v
o
l
u
t
i
o
n
A
u
t
o
E
n
c
o
d
e
r
(
C
A
E
)
Emp
i
r
i
c
a
l
M
o
d
e
D
e
c
o
mp
o
s
i
t
i
o
n
(
EM
D
)
9
9
%
9
4
%
1
0
0
%
F
i
t
n
e
ss,
W
a
l
k
i
n
g
3
5
0
7
C
N
N
[
1
8
]
9
0
.
4
2
%
Jo
g
g
i
n
g
,
W
a
l
k
i
n
g
,
U
p
s
a
t
r
a
i
r
,
D
o
w
n
s
t
a
i
r
,
S
i
t
t
i
n
g
,
S
t
a
n
d
i
n
g
,
La
y
i
n
g
2
0
0
8
C
N
N
+
K
N
N
[
1
9
]
C
N
N
+
S
V
M
7
1
%
9
4
%
1
5
mi
n
u
t
e
s
o
f
w
a
l
k
i
n
g
1
.
0
0
0
9
C
N
N
2
D
[
2
0
]
C
N
N
-
pf
(
p
a
r
t
i
a
l
f
u
l
l
)
C
N
N
-
p
f
f
(
p
a
r
t
i
a
l
f
u
l
l
w
e
i
g
h
t
)
7
9
.
7
3
%
6
6
.
6
5
%
9
9
.
6
6
%
S
t
a
n
d
i
n
g
s
t
i
l
l
,
si
t
t
i
n
g
a
n
d
r
e
l
a
x
i
n
g
,
l
a
y
i
n
g
d
o
w
n
,
w
a
l
k
i
n
g
,
c
l
i
m
b
i
n
g
st
a
i
r
s,
w
a
i
s
t
b
e
n
d
s
f
o
r
w
a
r
d
,
t
h
e
f
r
o
n
t
a
l
e
l
e
v
a
t
i
o
n
o
f
a
r
ms
,
k
n
e
e
s
b
e
n
d
i
n
g
,
c
y
c
l
i
n
g
,
j
o
g
g
i
n
g
,
r
u
n
n
i
n
g
,
j
u
mp
i
n
g
f
r
o
n
t
a
n
d
b
a
c
k
1
0
.
0
0
0
10
C
N
N
[
2
1
]
8
8
%
c
l
o
c
k
w
i
se
-
d
r
a
w
c
i
r
c
l
e
,
c
o
u
n
t
e
r
c
l
o
c
k
w
i
se
,
u
p
p
e
r
-
r
i
g
h
t
-
d
r
a
w
,
u
p
p
e
r
-
l
e
f
t
-
d
r
a
w
,
s
t
r
a
i
g
h
t
-
d
r
a
w
f
r
o
m l
e
f
t
t
o
r
i
g
h
t
,
st
r
a
i
g
h
t
-
d
r
a
w
f
r
o
m
r
i
g
h
t
t
o
l
e
f
t
.
2
0
,
0
0
0
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Hu
ma
n
a
ctivity
r
ec
o
g
n
itio
n
fo
r
s
ta
tic
a
n
d
d
yn
a
mic
a
ctivity
u
s
in
g
… (
A
g
u
s
E
ko
Min
a
r
n
o
)
1859
T
ab
le
2.
Dis
tr
ib
u
tio
n
o
f
ac
tiv
it
y
d
ata
C
a
t
e
g
o
r
y
A
c
t
i
v
i
t
y
Tr
a
i
n
Te
st
D
y
n
a
mi
c
W
a
l
k
i
n
g
W
a
l
k
i
n
g
U
p
s
t
a
i
r
W
a
l
k
i
n
g
D
o
w
n
st
a
i
r
1
2
2
6
1
0
7
3
9
8
7
4
9
6
4
7
1
4
2
0
S
t
a
t
ic
S
t
a
n
d
i
n
g
La
y
i
n
g
S
i
t
t
i
n
g
1
4
2
3
1
4
1
3
1
2
9
3
5
5
6
5
4
5
5
0
8
2.
M
E
T
H
O
D
2
.
1
.
D
a
t
a
s
et
UC
I
d
ataset
is
a
p
o
p
u
lar
d
ataset
u
s
ed
in
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
lear
n
in
g
[
2
2
]
.
T
h
is
d
ataset
was
o
b
tain
ed
f
r
o
m
3
0
v
o
lu
n
teer
s
wh
o
ca
r
r
ied
o
u
t
v
ar
io
u
s
ac
tiv
ities
b
y
wea
r
in
g
th
eir
wa
is
tlin
e
wh
ile
d
o
in
g
s
ix
ac
tiv
ities
(
s
tan
d
in
g
,
s
itti
n
g
,
la
y
in
g
,
walk
i
n
g
,
walk
i
n
g
d
o
wn
s
tair
,
walk
in
g
u
p
s
tair
s
)
Fig
u
r
e
1
.
T
h
e
g
ad
g
ets
u
s
ed
in
d
ata
co
llectio
n
ca
n
r
ec
o
r
d
t
h
e
ac
tiv
ity
u
s
in
g
th
e
h
elp
o
f
t
h
e
ac
ce
ler
o
m
eter
an
d
g
y
r
o
s
co
p
e
s
en
s
o
r
s
th
at
h
av
e
b
ee
n
in
s
t
alled
in
th
e
g
a
d
g
et,
o
n
th
e
s
id
e
o
f
th
e
ac
ce
ler
o
m
ete
r
an
d
g
y
r
o
s
co
p
e
d
ata
co
llectio
n
p
r
o
ce
s
s
r
ec
o
r
d
ed
u
s
in
g
th
e
v
id
e
o
to
lab
el
it
m
a
n
u
ally
.
B
y
u
s
in
g
th
e
g
y
r
o
s
co
p
e
an
d
ac
ce
ler
o
m
eter
s
en
s
o
r
s
i
n
th
e
g
a
d
g
et,
th
e
y
o
b
tain
th
r
ee
-
ax
is
lin
ea
r
ac
ce
l
er
atio
n
(
XYZ
)
d
ata
f
r
o
m
th
e
ac
ce
ler
o
m
eter
s
en
s
o
r
,
an
d
th
e
g
y
r
o
s
co
p
e
s
en
s
o
r
g
en
er
ates
th
r
ee
-
a
x
is
an
g
u
lar
v
elo
city
(
XYZ
)
.
T
h
e
s
en
s
o
r
s
ig
n
al
is
th
en
p
r
o
ce
s
s
ed
u
s
in
g
n
o
i
s
e
f
ilter
s
an
d
th
en
in
th
e
s
am
p
le
in
f
ix
ed
co
n
tain
e
r
s
(
s
lid
in
g
win
d
o
ws)
at
in
ter
v
als
o
f
2
.
5
6
s
ec
o
n
d
s
with
an
o
v
er
lap
o
f
5
0
%.
T
h
e
p
r
o
ce
s
s
ed
d
ataset
is
d
iv
id
ed
in
to
7
0
% a
s
tr
ain
in
g
d
at
a,
an
d
3
0
% o
f
test
d
ata
is
s
h
o
wn
in
T
a
b
le
2
.
Sig
n
al
d
ata
f
r
o
m
d
y
n
am
ic
an
d
s
tatic
ac
tiv
ities
h
as
a
v
er
y
s
ig
n
if
ican
t
d
if
f
e
r
en
ce
,
as
s
ee
n
i
n
Fig
u
r
e
2
with
6
s
tatic
an
d
d
y
n
am
ic
ac
tiv
ities
.
Fig
u
r
e
3
(
a)
s
h
o
ws
th
at
th
er
e
is
a
p
r
o
b
lem
th
at
o
cc
u
r
s
in
th
e
HAR
th
at
i
s
th
e
s
im
ilar
ity
o
f
s
tatic
s
ig
n
al
d
ata
with
s
tan
d
in
g
an
d
s
itti
n
g
ac
tiv
ities
.
T
h
e
s
im
ilar
ities
o
f
s
tan
d
in
g
an
d
s
itti
n
g
ac
tiv
ity
d
ata
r
esu
lt
in
d
ee
p
lea
r
n
in
g
e
r
r
o
r
s
in
class
if
y
in
g
ac
tiv
ities
,
an
d
th
is
ca
n
lo
wer
t
h
e
l
ev
el
o
f
ac
cu
r
ac
y
in
th
e
o
v
er
all
HAR
p
r
o
ce
s
s
in
g
.
I
n
th
is
ar
ticle
,
we
u
s
e
t
-
SNE
,
wh
ich
ca
n
d
is
p
lay
a
h
ig
h
-
d
im
en
s
io
n
al
d
ata
s
p
r
ea
d
b
y
r
e
d
u
cin
g
its
d
im
e
n
s
io
n
ality
to
two
d
im
en
s
io
n
s
.
W
e
u
s
e
c
o
n
f
ig
u
r
atio
n
1
0
0
0
iter
atio
n
s
a
n
d
p
e
r
p
lex
ities
2
,
5
,
1
0
,
2
0
,
an
d
5
0
F
ig
u
r
e
3
(
b
)
.
F
ig
u
r
e
3
(
b
)
r
esu
lt
t
-
SNE
with
p
er
p
lex
ity
2
a
n
d
5
u
s
in
g
1
0
0
0
iter
atio
n
s
h
o
ws
all
g
r
o
u
p
ac
tiv
ity
with
th
e
s
am
e
ty
p
e
o
f
d
ata,
b
u
t
th
er
e
ar
e
s
tan
d
in
g
an
d
s
itti
n
g
ac
tiv
ities
th
at
h
av
e
th
e
s
am
e
ty
p
e
o
f
d
ata.
Fig
u
r
e
1
.
V
o
l
u
n
teer
ac
tiv
ity
d
ata
f
o
r
s
tan
d
in
g
,
s
itti
n
g
,
lay
in
g
,
walk
in
g
,
walk
in
g
d
o
w
n
s
tair
s
,
walk
in
g
2
.
2
.
H
y
perpa
ra
m
e
t
er
H
y
p
e
r
p
a
r
a
m
e
t
e
r
i
s
a
m
e
t
h
o
d
i
n
t
h
e
n
e
u
r
a
l
n
e
t
w
o
r
k
t
h
a
t
a
l
l
o
w
s
u
s
e
r
s
t
o
o
b
t
a
i
n
a
c
o
m
b
i
n
a
t
i
o
n
o
f
p
a
r
a
m
e
t
e
r
s
t
h
a
t
h
a
v
e
t
h
e
b
e
s
t
a
c
c
u
r
a
c
y
v
a
l
u
e
f
r
o
m
a
n
u
m
b
e
r
o
f
p
r
e
v
i
o
u
s
n
e
u
r
a
l
n
e
t
w
o
r
k
c
o
m
p
u
t
i
n
g
s
t
e
p
s
[
2
3
]
.
T
h
e
co
m
b
in
atio
n
o
f
p
ar
a
m
eter
s
o
b
tain
ed
b
y
u
s
in
g
h
y
p
er
p
a
r
am
eter
in
clu
d
e
s
th
e
n
u
m
b
er
o
f
lay
er
s
u
s
ed
,
th
e
m
ap
p
in
g
f
ea
tu
r
e
,
s
ize
co
n
v
o
lu
tio
n
f
ilt
er
,
s
ize
p
o
o
lin
g
d
ata
s
et
[
2
4
]
–
[
3
0
]
.
T
h
e
p
ar
am
eter
s
u
s
ed
o
n
th
e
p
r
o
p
o
s
ed
C
NN
m
o
d
el
b
ef
o
r
e
t
u
n
in
g
wer
e
s
ee
n
in
T
a
b
le
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
6
,
Dec
em
b
e
r
2
0
2
1
:
18
57
-
18
64
1860
Fig
u
r
e
2
.
Static a
n
d
d
y
n
a
m
ic
a
ctiv
ity
: stan
d
in
g
,
s
itti
n
g
,
lay
in
g
,
walk
in
g
,
walk
i
n
g
d
o
wn
s
tair
s
,
an
d
walk
in
g
u
p
s
tair
s
(
a)
(
b
)
Fig
u
r
e
3
.
Sp
r
ea
d
d
ata
s
tan
d
i
n
g
,
s
itti
n
g
,
lay
in
g
,
walk
in
g
,
walk
in
g
d
o
wn
s
tair
s
,
walk
in
g
u
p
s
tair
s
:
(
a)
d
ata
s
im
ilar
ity
s
tan
d
in
g
an
d
s
itti
n
g
ac
tiv
ities
r
esu
lt in
d
ee
p
lear
n
in
g
er
r
o
r
s
in
class
if
y
in
g
ac
tiv
ity
an
d
(
b
)
co
n
f
ig
u
r
atio
n
1
0
0
0
iter
atio
n
s
an
d
p
er
p
lex
ities
2
,
5
,
1
0
,
2
0
,
an
d
5
0
,
s
tan
d
in
g
an
d
s
itti
n
g
a
ctiv
ities
th
at
h
av
e
th
e
s
am
e
ty
p
e
o
f
d
ata
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Hu
ma
n
a
ctivity
r
ec
o
g
n
itio
n
fo
r
s
ta
tic
a
n
d
d
yn
a
mic
a
ctivity
u
s
in
g
… (
A
g
u
s
E
ko
Min
a
r
n
o
)
1861
T
ab
le
3.
C
NN
u
s
ed
in
2
class
La
y
e
r
P
a
r
a
me
t
e
r
S
c
o
r
e
La
y
e
r
1
F
i
l
t
e
r
K
e
r
n
e
l
si
z
e
A
c
t
i
v
a
t
i
o
n
32
3
Re
L
u
k
e
r
n
e
l
_
i
n
i
t
i
a
l
i
z
e
r
h
e
_
u
n
i
f
o
r
m
i
n
p
u
t
_
sh
a
p
e
(
1
2
8
,
9
)
La
y
e
r
2
F
i
l
t
e
r
K
e
r
n
e
l
si
z
e
A
c
t
i
v
a
t
i
o
n
32
3
Re
L
u
k
e
r
n
e
l
_
i
n
i
t
i
a
l
i
z
e
r
h
e
_
u
n
i
f
o
r
m
D
r
o
p
o
u
t
u
n
i
f
o
r
m
0
.
6
M
a
x
P
o
o
l
i
n
g
1
D
p
o
o
l
_
s
i
z
e
2
F
l
a
t
t
e
n
F
l
a
t
t
e
n
D
e
n
se
A
c
t
:
R
e
L
u
50
D
e
n
se
A
c
t
:
so
f
t
ma
x
2
K
e
r
a
s.
o
p
t
i
m
i
z
e
r
s
A
d
a
m
0
.
0
0
1
n
b
_
e
p
o
c
h
1
0
0
3.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
E
v
alu
atio
n
o
f
p
r
ed
icted
r
esu
lts
f
r
o
m
ea
c
h
m
o
d
el
u
s
in
g
a
c
o
n
f
u
s
io
n
m
atr
ix
.
T
h
e
c
o
n
f
u
s
io
n
m
atr
ix
is
a
m
eth
o
d
u
s
ed
to
p
er
f
o
r
m
ac
c
u
r
ac
y
ca
lcu
latio
n
s
o
n
a
p
r
ed
ictiv
e
s
y
s
tem
.
C
o
n
f
u
s
io
n
m
at
r
ix
co
n
tain
s
ac
t
u
al
in
f
o
r
m
atio
n
an
d
p
r
ed
ictio
n
s
o
n
th
e
class
if
icatio
n
s
y
s
t
em
.
T
o
f
in
d
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
an
d
r
ec
all
s
eq
u
en
tially
u
s
in
g
(
3
)
-
(
5
)
.
=
∑
=
1
+
+
T
+
+
∗
100%
(
3
)
=
∑
=
1
∑
(
+
F
)
=
1
∗
100%
(
4
)
=
∑
=
1
∑
(
+
FN
)
=
1
∗
100%
(
5
)
I
n
th
is
a
r
ticle
,
th
e
au
th
o
r
p
e
r
f
o
r
m
s
test
in
g
u
s
in
g
C
NN
o
n
t
h
e
HAR
d
ataset
in
to
2
class
es
,
n
am
ely
th
e
d
y
n
a
m
ic
class
an
d
th
e
s
tatic
c
lass
with
th
e
p
ar
am
eter
s
s
h
o
wn
in
T
ab
le
4
.
T
h
e
u
s
e
o
f
h
y
p
er
p
ar
am
eter
tu
n
in
g
to
g
et
th
e
b
est
p
ar
am
eter
co
m
b
in
atio
n
s
g
en
e
r
ates
th
e
h
ig
h
est
ac
cu
r
ac
y
o
n
ea
ch
s
tatic
an
d
d
y
n
am
ic
d
ataset
v
iewa
b
le
in
T
ab
le
5
.
H
y
p
er
p
a
r
am
eter
ca
n
p
r
o
v
i
d
e
th
e
co
n
f
ig
u
r
atio
n
o
f
th
e
p
ar
am
ete
r
s
n
ee
d
ed
f
o
r
C
NN
m
o
d
els
o
f
th
e
s
elec
ted
d
ataset
b
y
r
a
n
d
o
m
l
y
cr
ea
tin
g
a
co
m
b
in
atio
n
o
f
p
a
r
am
eter
s
.
On
th
e
f
ir
s
t
lay
er
,
th
e
h
y
p
er
p
ar
am
eter
will
s
elec
t
th
e
f
ilter
v
alu
es
b
etwe
en
(
2
8
,
4
3
,
o
r
4
2
)
,
s
im
ilar
ly
th
e
v
alu
es
o
f
t
h
e
k
er
n
el
s
ize,
m
ax
p
o
o
lin
g
id
,
b
atc
h
s
ize,
e
p
o
c
h
,
an
d
d
e
n
s
e
p
ar
am
eter
s
.
W
h
ile
th
e
co
n
f
i
g
u
r
atio
n
o
f
t
h
e
d
r
o
p
o
u
t
p
ar
am
eter
wi
ll
b
e
d
eter
m
in
ed
u
s
in
g
a
v
alu
e
b
et
wee
n
0
.
4
5
-
0
.
7
.
T
h
e
o
p
tim
izer
will
b
e
u
s
ed
b
etwe
en
Ad
am
a
n
d
R
MSp
r
o
p
with
a
v
alu
e
b
etwe
en
0
.
0
0
0
6
5
-
0
.
0
0
4
.
Hy
p
er
p
ar
a
m
eter
t
u
n
in
g
is
ex
ec
u
ted
b
y
th
e
n
u
m
b
er
o
f
m
o
d
el
s
to
b
e
g
en
er
ated
as
m
an
y
as
1
0
0
m
o
d
els.
T
h
e
o
v
e
r
all
co
n
f
ig
u
r
at
io
n
ca
n
b
e
s
ee
n
in
T
ab
le
4
.
1
0
0
co
m
b
in
atio
n
s
ar
e
ex
ec
u
ted
u
s
in
g
Hy
p
er
p
ar
a
m
eter
tu
n
in
g
,
SA
d
a
tasets
g
et
an
ac
cu
r
ac
y
o
f
9
7
%
in
d
ata
tr
ain
a
n
d
9
6
%
in
th
e
v
al
id
atio
n
d
ata
s
h
o
w
n
in
Fig
u
r
e
4
,
wh
ile
th
e
ac
cu
r
ac
y
o
f
th
e
DA
Data
s
et
g
en
er
ates
an
ac
cu
r
ac
y
v
al
u
e
o
f
1
0
0
%
o
n
th
e
d
ata
tr
ain
an
d
9
7
.
4
% in
th
e
v
alid
atio
n
d
ata
s
h
o
wn
in
Fig
u
r
e
5
.
T
ab
le
4.
C
NN
tu
n
in
g
h
y
p
e
r
p
ar
am
eter
p
r
ep
a
r
atio
n
La
y
e
r
P
a
r
a
me
t
e
r
S
c
o
r
e
A
n
n
o
t
a
t
i
o
n
La
y
e
r
P
a
r
a
me
t
e
r
S
c
o
r
e
A
n
n
o
t
a
t
i
o
n
La
y
e
r
1
F
i
l
t
e
r
K
e
r
n
e
l
si
z
e
A
c
t
i
v
a
t
i
o
n
2
8
,
3
2
,
4
2
3
,
5
,
7
R
e
l
u
C
h
o
i
c
e
C
h
o
i
c
e
F
i
x
e
d
D
r
o
p
o
u
t
u
n
i
f
o
r
m
0
.
4
5
-
0
.
7
R
a
n
g
e
k
e
r
n
e
l
_
i
n
i
t
i
a
l
i
z
e
r
h
e
_
u
n
i
f
o
r
m
F
i
x
e
d
M
a
x
P
o
o
l
i
n
g
1
D
p
o
o
l
_
s
i
z
e
2
,
3
,
5
C
g
o
i
c
e
k
e
r
n
e
l
_
r
e
g
u
l
a
r
i
z
e
r
l2
0
,
3
F
i
x
e
d
F
l
a
t
t
e
n
F
l
a
t
t
e
n
F
i
x
e
d
i
n
p
u
t
_
sh
a
p
e
(
1
2
8
,
9
)
F
i
d
e
d
D
e
n
se
A
c
t
:
r
e
l
u
1
6
,
3
2
,
6
4
C
h
o
i
c
e
La
y
e
r
2
F
i
l
t
e
r
K
e
r
n
e
l
si
z
e
A
c
t
i
v
a
t
i
o
n
1
6
,
2
4
,
4
2
3
,
5
,
7
R
e
l
u
C
h
o
i
c
e
C
h
o
i
c
e
F
i
x
e
d
D
e
n
se
A
c
t
:
so
f
t
ma
x
3
F
i
x
e
d
k
e
r
n
e
l
_
i
n
i
t
i
a
l
i
z
e
r
h
e
_
u
n
i
f
o
r
m
F
i
x
e
d
k
e
r
a
s
.
o
p
t
i
m
i
z
e
r
s
A
d
a
m,
r
msp
r
o
p
0
.
0
0
0
6
5
-
0
.
0
0
4
R
a
n
g
e
k
e
r
n
e
l
_
r
e
g
u
l
a
r
i
z
e
r
l2
(
0
,
2
)
F
i
x
e
d
b
a
t
c
h
_
s
i
z
e
1
6
,
3
2
,
6
4
C
h
o
i
c
e
n
b
_
e
p
o
c
h
3
5
,
4
0
C
h
o
i
c
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
6
,
Dec
em
b
e
r
2
0
2
1
:
18
57
-
18
64
1862
Fig
u
r
e
4
.
Acc
u
r
ac
y
a
n
d
lo
s
s
SA
Fig
u
r
e
5
.
Acc
u
r
ac
y
a
n
d
lo
s
s
DA
Mo
d
els
o
b
tain
e
d
f
r
o
m
th
r
ee
d
atasets
ar
e
c
o
m
b
in
e
d
in
t
o
th
e
m
ain
m
o
d
el
u
s
in
g
th
e
d
iv
id
e
a
n
d
co
n
q
u
e
r
m
eth
o
d
s
th
at
ca
n
i
d
en
tify
all
h
u
m
a
n
ac
tiv
ities
(
walk
in
g
,
walk
in
g
u
p
,
walk
in
g
d
o
w
n
,
s
tan
d
in
g
,
lay
in
g
,
an
d
s
itti
n
g
)
.
T
h
is
m
o
d
el
g
en
e
r
ates
an
ac
cu
r
ac
y
o
f
9
8
.
3
%
in
d
ata
tr
ain
an
d
9
7
%
in
v
alid
ati
o
n
d
ata
.
T
h
e
m
o
d
el
ar
ch
itectu
r
e
I
S
s
ee
n
in
Fig
u
r
e
6
.
T
h
e
co
m
p
a
r
is
o
n
o
f
th
is
s
tu
d
y
with
p
r
ev
io
u
s
r
esear
ch
titl
ed
"Hu
m
an
ac
tiv
ity
r
ec
o
g
n
itio
n
with
s
m
ar
tp
h
o
n
e
s
en
s
o
r
s
u
s
in
g
d
ee
p
lear
n
in
g
n
e
u
r
al
n
etwo
r
k
s
"
[
1
4
]
us
in
g
th
e
s
am
e
d
ataset
as
well
as
tu
n
in
g
o
f
t
h
e
lear
n
in
g
r
ate
=0
.
0
0
6
r
esu
lted
in
an
ac
c
u
r
ac
y
o
f
9
3
.
7
5
%.
I
n
co
m
p
ar
is
o
n
,
th
e
p
r
o
p
o
s
ed
C
NN
m
o
d
el
ac
h
iev
e
d
a
9
7
% a
cc
u
r
a
cy
ca
n
b
e
s
ee
n
in
T
ab
le
6
.
T
ab
le
5
.
C
o
n
f
u
s
io
n
m
atr
ix
C
NN
r
esear
ch
r
esu
lts
in
h
y
p
e
r
p
ar
am
eter
u
s
e
d
iv
id
e
a
n
d
c
o
n
q
u
er
P
r
e
d
i
c
t
e
d
c
l
a
ss
R
e
c
a
l
l
W
WU
WD
Si
St
L
A
c
t
u
a
l
C
l
a
ss
W
a
l
k
i
n
g
4
9
4
0
0
3
0
0
9
9
.
3
%
W
.
U
p
0
4
4
7
0
2
0
0
9
9
.
5
%
W
.
D
o
w
n
2
24
4
2
0
0
0
0
9
4
.
2
%
S
i
t
t
i
n
g
0
0
0
4
6
0
29
0
9
4
.
1
%
S
t
a
n
d
i
n
g
0
0
0
22
5
0
1
0
9
5
.
8
%
La
y
i
n
g
0
0
0
4
2
5
3
7
9
8
.
9
P
r
e
d
i
c
t
i
o
n
9
9
.
5
%
9
5
%
1
0
0
%
9
4
%
9
4
%
1
0
0
%
9
7
%
T
ab
le
6
.
C
o
m
p
a
r
is
o
n
o
f
p
r
e
v
io
u
s
s
tu
d
ies
M
e
t
h
o
d
A
c
c
u
r
a
c
y
H
C
F
+
N
B
[
1
4
]
7
9
.
4
3
%
H
C
F
+
J4
8
[
1
4
]
8
2
.
6
2
%
H
C
F
+
A
N
N
[
1
4
]
8
2
.
2
7
%
H
C
F
+
S
V
M
[
1
4
]
7
7
.
6
6
%
C
o
n
v
n
e
t
[
1
4
]
9
3
.
7
5
%
CN
N
P
r
o
p
o
s
e
d
me
t
h
o
d
9
7
%
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Hu
ma
n
a
ctivity
r
ec
o
g
n
itio
n
fo
r
s
ta
tic
a
n
d
d
yn
a
mic
a
ctivity
u
s
in
g
… (
A
g
u
s
E
ko
Min
a
r
n
o
)
1863
Fig
u
r
e
6
.
SA a
n
d
DA
co
m
b
i
n
e
d
m
o
d
els
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
p
r
o
p
o
s
ed
a
h
u
m
a
n
ac
tiv
ity
r
ec
o
g
n
itio
n
u
s
in
g
C
NN
with
6
class
e
s
(
walk
in
g
,
walk
in
g
u
p
s
tair
s
,
walk
in
g
d
o
wn
s
tair
,
s
itti
n
g
,
s
tan
d
in
g
,
an
d
lay
in
g
)
.
B
ased
o
n
th
e
r
esu
lts
an
d
d
is
cu
s
s
io
n
,
th
e
d
i
v
id
e
an
d
co
n
q
u
e
r
m
eth
o
d
co
m
b
in
e
wit
h
C
NN
tu
n
in
g
h
y
p
er
p
ar
am
et
er
in
ea
ch
ca
te
g
o
r
y
o
f
d
atasets
ac
h
iev
ed
9
7
%
ac
cu
r
ac
y
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
e
l
also
s
o
lv
e
d
th
e
s
im
ilar
ity
p
r
o
b
lem
s
with
co
n
f
u
s
in
g
s
tatic
(
s
itti
n
g
an
d
s
tan
d
in
g
)
d
atasets
.
T
h
e
h
ig
h
est
ac
cu
r
a
cy
r
ea
ch
e
d
1
0
0
%
in
t
h
e
s
tu
d
y
was
s
h
o
wn
o
n
th
e
ac
t
iv
ities
o
f
w
alk
in
g
d
o
wn
s
tair
an
d
l
a
y
in
g
,
wh
ile
t
h
e
lo
west a
cc
u
r
ac
y
o
f
9
4
% is
o
wn
ed
b
y
t
h
e
a
ctiv
ities
o
f
s
itti
n
g
an
d
s
tan
d
i
n
g
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
an
k
y
o
u
t
o
th
e
I
n
f
o
r
m
atics
Dep
ar
tm
en
t
an
d
I
n
f
o
r
m
atics
L
ab
o
r
ato
r
y
,
Un
iv
e
r
s
itas
Mu
h
am
m
ad
iy
ah
Ma
lan
g
wh
ich
h
as
b
ec
o
m
e
a
p
lace
f
o
r
r
esear
ch
er
s
to
d
ev
elo
p
th
is
jo
u
r
n
al
r
esear
ch
.
Ho
p
ef
u
ll
y
,
th
is
r
esear
ch
ca
n
m
ak
e
a
m
ajo
r
co
n
tr
i
b
u
tio
n
t
o
t
h
e
ad
v
a
n
ce
m
en
t o
f
tech
n
o
lo
g
y
in
I
n
d
o
n
esia.
RE
F
E
R
E
NC
E
S
[1
]
L.
Ch
e
n
a
n
d
C.
Nu
g
e
n
t,
"
Hu
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
i
ti
o
n
a
n
d
b
e
h
a
v
io
u
r
a
n
a
ly
sis
,
"
S
p
ri
n
g
e
r,
2
0
1
9
.
[2
]
Y.
Xia
,
J.
Zh
a
n
g
,
Q.
Ye
,
N.
C
h
e
n
g
,
a
n
d
Y.
L
u
,
“
E
v
a
lu
a
ti
o
n
o
f
d
e
e
p
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s
fo
r
d
e
tec
ti
o
n
o
f
fre
e
z
in
g
o
f
g
a
it
in
P
a
rk
i
n
so
n
’s
d
i
se
a
se
p
a
t
ien
ts,”
Bi
o
me
d
ica
l
S
ig
n
a
l
Pro
c
e
ss
in
g
a
n
d
C
o
n
tro
l
,
v
o
l.
4
6
,
p
p
.
2
2
1
–
2
3
0
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/
j.
b
s
p
c
.
2
0
1
8
.
0
7
.
0
1
5
.
[3
]
A.
E.
M
i
n
a
rn
o
,
W.
Ku
su
m
a
,
a
n
d
H.
Wi
b
o
w
o
,
“
P
e
rfo
rm
a
n
c
e
Co
m
p
a
risso
n
Ac
ti
v
it
y
Re
c
o
g
n
it
io
n
u
sin
g
L
o
g
ist
ic
Re
g
re
ss
io
n
a
n
d
S
u
p
p
o
r
t
Ve
c
to
r
M
a
c
h
in
e
,
”
2
0
2
0
3
rd
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
In
telli
g
e
n
t
Au
to
n
o
mo
u
s
S
y
ste
ms
(ICo
IAS
)
,
2
0
2
0
,
p
p
.
1
9
–
2
4
,
d
o
i:
1
0
.
1
1
0
9
/ICo
IA
S
4
9
3
1
2
.
2
0
2
0
.
9
0
8
1
8
5
8
.
[4
]
A.
M
in
a
r
n
o
,
W.
Ku
su
m
a
,
H.
Wi
b
o
wo
,
D
.
Ak
b
i,
a
n
d
N.
Ja
wa
s,
“
S
in
g
le
tri
a
x
ial
a
c
c
e
lero
m
e
ter
-
g
y
ro
sc
o
p
e
c
las
sifica
ti
o
n
fo
r
h
u
m
a
n
a
c
ti
v
i
ty
re
c
o
g
n
it
io
n
,
”
2
0
2
0
8
t
h
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
I
n
f
o
rm
a
ti
o
n
a
n
d
C
o
mm
u
n
ic
a
ti
o
n
T
e
c
h
n
o
lo
g
y
(ICo
ICT
)
,
2
0
2
0
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/IC
o
ICT4
9
3
4
5
.
2
0
2
0
.
9
1
6
6
3
2
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
6
,
Dec
em
b
e
r
2
0
2
1
:
18
57
-
18
64
1864
[5
]
C.
Jo
b
a
n
p
u
tra,
J.
Ba
v
is
h
i,
a
n
d
N.
Do
sh
i,
“
Hu
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
i
o
n
:
A
su
rv
e
y
,
”
Pro
c
e
d
ia
Co
mp
u
ter
S
c
ien
c
e
,
v
o
l.
1
5
5
,
p
p
.
6
9
8
–
7
0
3
,
2
0
1
9
,
d
o
i:
1
0
.
1
0
1
6
/
j.
p
r
o
c
s.2
0
1
9
.
0
8
.
1
0
0
.
[6
]
F
.
F
o
e
rste
r,
M
.
S
m
e
ja,
a
n
d
J.
F
a
h
re
n
b
e
rg
,
“
De
tec
ti
o
n
o
f
p
o
stu
re
a
n
d
m
o
t
io
n
b
y
a
c
c
e
lero
m
e
try
:
a
v
a
li
d
a
ti
o
n
st
u
d
y
i
n
a
m
b
u
lato
ry
m
o
n
it
o
rin
g
,
”
Co
m
p
u
t
e
rs
in
h
u
ma
n
b
e
h
a
v
io
r
,
v
o
l.
1
5
,
n
o
.
5
,
p
p
.
5
7
1
–
5
8
3
,
1
9
9
9
,
d
o
i:
1
0
.
1
0
1
6
/S
0
7
4
7
-
5
6
3
2
(
9
9
)
0
0
0
3
7
-
0.
[7
]
L.
Ba
o
a
n
d
S
.
S
.
In
ti
l
le,
“
Ac
ti
v
it
y
re
c
o
g
n
i
ti
o
n
fro
m
u
se
r
-
a
n
n
o
tate
d
a
c
c
e
lera
ti
o
n
d
a
ta,”
In
ter
n
a
ti
o
n
a
l
c
o
n
fer
e
n
c
e
o
n
p
e
rv
a
siv
e
c
o
mp
u
ti
n
g
,
v
o
l.
3
0
0
1
,
2
0
0
4
,
p
p
.
1
–
1
7
,
d
o
i:
1
0
.
1
0
0
7
/9
7
8
-
3
-
5
4
0
-
2
4
6
4
6
-
6
_
1
.
[8
]
T.
P
l
ö
tz,
N.
Ha
m
m
e
rla,
a
n
d
P
.
Ol
iv
ier,
“
F
e
a
tu
re
lea
rn
i
n
g
fo
r
a
c
ti
v
it
y
re
c
o
g
n
it
i
o
n
i
n
u
b
i
q
u
it
o
u
s
c
o
m
p
u
ti
n
g
,
”
T
we
n
ty
-
se
c
o
n
d
i
n
ter
n
a
ti
o
n
a
l
j
o
in
t
c
o
n
fe
re
n
c
e
o
n
a
rtif
icia
l
i
n
telli
g
e
n
c
e
,
2
0
1
1
,
d
o
i:
1
0
.
5
5
9
1
/
9
7
8
-
1
-
5
7
7
3
5
-
5
1
6
-
8
/IJCAI1
1
-
2
9
0
.
[9
]
T.
G
o
n
z
a
lez
,
"
Ha
n
d
b
o
o
k
o
f
a
p
p
ro
x
ima
ti
o
n
a
lg
o
r
it
h
m
s a
n
d
m
e
tah
e
u
risti
c
s
,"
CRC
Pre
ss
,
2
0
0
7
.
[1
0
]
O.
C.
K
u
rb
a
n
a
n
d
T.
Yıl
d
ırı
m
,
“
Da
il
y
m
o
ti
o
n
re
c
o
g
n
it
i
o
n
sy
ste
m
b
y
a
tr
iax
i
a
l
a
c
c
e
lero
m
e
ter
u
sa
b
le
in
d
iffere
n
t
p
o
siti
o
n
s,”
I
EE
E
S
e
n
so
rs
J
o
u
rn
a
l
,
v
o
l.
1
9
,
n
o
.
1
7
,
p
p
.
7
5
4
3
–
7
5
5
2
,
2
0
1
9
,
d
o
i:
1
0
.
1
1
0
9
/J
S
EN.
2
0
1
9
.
2
9
1
5
5
2
4
.
[1
1
]
M
.
Y
a
n
g
,
H
.
Z
h
e
n
g
,
H
.
W
a
n
g
,
a
n
d
S
.
M
c
C
l
e
a
n
,
“
iG
A
I
T
:
a
n
i
n
t
e
ra
c
t
i
v
e
a
c
c
e
l
e
r
o
m
e
te
r
b
a
se
d
g
a
i
t
a
n
a
l
y
s
i
s
s
y
s
t
e
m
,
”
C
o
m
p
u
t
e
r
m
e
t
h
o
d
s
a
n
d
p
r
o
g
r
a
m
s
i
n
b
i
o
m
e
d
i
c
i
n
e
,
v
o
l
.
1
0
8
,
n
o
.
2
,
p
p
.
7
1
5
–
7
2
3
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
m
p
b
.
2
0
1
2
.
0
4
.
0
0
4
.
[1
2
]
D
.
F
u
e
n
t
e
s
,
L
.
G
o
n
z
a
le
z
-
Ab
r
i
l
,
A
n
g
u
l
o
.
C
,
a
n
d
J
.
O
r
t
e
g
a
,
“
O
n
l
i
n
e
m
o
ti
o
n
r
e
c
o
g
n
i
t
i
o
n
u
s
i
n
g
a
n
a
c
c
e
l
e
r
o
m
e
t
e
r
i
n
a
m
o
b
i
l
e
d
e
v
i
c
e
,
”
E
x
p
e
r
t
sy
s
t
e
m
s
w
i
t
h
a
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
9
,
n
o
.
3
,
p
p
.
2
4
6
1
–
2
4
6
5
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
s
w
a
.
2
0
1
1
.
0
8
.
0
9
8
.
[1
3
]
A
.
M
.
Kh
a
n
,
“
Re
c
o
g
n
izi
n
g
p
h
y
sic
a
l
a
c
ti
v
it
ies
u
sin
g
Wi
i
re
m
o
te,”
In
t
e
rn
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
In
fo
rm
a
t
io
n
a
n
d
Ed
u
c
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
3
,
n
o
.
1
,
2
0
1
3
,
d
o
i:
1
0
.
7
7
6
3
/IJIE
T.
2
0
1
3
.
V3
.
2
3
4
.
[1
4
]
C.
Ro
n
a
o
a
n
d
S
.
Ch
o
,
“
Hu
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
io
n
with
sm
a
rtp
h
o
n
e
se
n
so
rs
u
si
n
g
d
e
e
p
lea
rn
in
g
n
e
u
ra
l
n
e
two
rk
s,”
Exp
e
rt sy
ste
ms
wit
h
a
p
p
li
c
a
ti
o
n
s
,
v
o
l.
5
9
,
p
p
.
2
3
5
–
2
4
4
,
2
0
1
6
,
d
o
i:
1
0
.
1
0
1
6
/j
.
e
sw
a
.
2
0
1
6
.
0
4
.
0
3
2
.
[1
5
]
W.
Qi,
H.
S
u
,
C.
Ya
n
g
,
G
.
F
e
rri
g
n
o
,
E
.
De
M
o
m
i,
a
n
d
A.
Al
iv
e
r
ti
,
“
A
fa
st
a
n
d
ro
b
u
st
d
e
e
p
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s
f
o
r
c
o
m
p
lex
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
io
n
u
sin
g
s
m
a
rtp
h
o
n
e
,
”
S
e
n
so
rs
,
v
o
l.
1
9
,
n
o
.
1
7
,
2
0
1
9
,
d
o
i:
1
0
.
3
3
9
0
/s1
9
1
7
3
7
3
1
.
[1
6
]
J.
Nu
n
e
z
,
R.
Ca
b
id
o
,
J.
P
a
n
tri
g
o
,
A.
M
o
n
tem
a
y
o
r,
a
n
d
J.
Ve
lez
,
“
Co
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
r
k
s
a
n
d
lo
n
g
sh
o
r
t
-
term
m
e
m
o
ry
fo
r
sk
e
leto
n
-
b
a
se
d
h
u
m
a
n
a
c
ti
v
it
y
a
n
d
h
a
n
d
g
e
stu
re
re
c
o
g
n
it
io
n
,
”
P
a
tt
e
rn
Rec
o
g
n
it
io
n
,
v
o
l.
7
6
,
p
p
.
8
0
–
9
4
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/
j.
p
a
tco
g
.
2
0
1
7
.
1
0
.
0
3
3
.
[1
7
]
A.
Ig
n
a
to
v
,
“
Re
a
l
-
ti
m
e
h
u
m
a
n
a
c
t
iv
it
y
re
c
o
g
n
it
i
o
n
fr
o
m
a
c
c
e
lero
m
e
ter
d
a
ta
u
si
n
g
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
two
r
k
s,”
Ap
p
li
e
d
S
o
ft
Co
mp
u
ti
n
g
,
v
o
l.
6
2
,
p
p
.
9
1
5
–
9
2
2
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/j
.
a
so
c
.
2
0
1
7
.
0
9
.
0
2
7
.
[1
8
]
M
.
G
a
d
a
leta
a
n
d
M
.
R
o
ss
i,
“
I
DN
e
t:
S
m
a
rtp
h
o
n
e
-
b
a
se
d
g
a
it
re
c
o
g
n
it
io
n
with
c
o
n
v
o
lu
t
io
n
a
l
n
e
u
ra
l
n
e
two
rk
s,”
P
a
tt
e
r
n
Rec
o
g
n
it
io
n
,
v
o
l.
7
4
,
p
p
.
2
5
–
3
7
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/j
.
p
a
tco
g
.
2
0
1
7
.
0
9
.
0
0
5
.
[1
9
]
S
.
Ha
a
n
d
S
.
Ch
o
i,
“
Co
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
r
k
s f
o
r
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
io
n
u
sin
g
m
u
lt
ip
le ac
c
e
lero
m
e
ter an
d
g
y
r
o
sc
o
p
e
se
n
s
o
rs,”
2
0
1
6
I
n
ter
n
a
ti
o
n
a
l
J
o
in
t
Co
n
fer
e
n
c
e
o
n
Ne
u
ra
l
Ne
two
rk
s
(IJ
CNN)
,
2
0
1
6
,
p
p
.
3
8
1
–
3
8
8
,
d
o
i:
1
0
.
1
1
0
9
/IJCNN
.
2
0
1
6
.
7
7
2
7
2
2
4
.
[2
0
]
J.
Kim
,
G
.
Ho
n
g
,
B.
Kim
,
a
n
d
D
.
Do
g
ra
,
“
d
e
e
p
G
e
stu
re
:
De
e
p
lea
rn
i
n
g
-
b
a
se
d
g
e
stu
re
re
c
o
g
n
it
i
o
n
sc
h
e
m
e
u
sin
g
m
o
ti
o
n
se
n
so
rs,”
Disp
la
y
s
,
v
o
l
.
5
5
,
p
p
.
3
8
–
4
5
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/j
.
d
isp
la.
2
0
1
8
.
0
8
.
0
0
1
.
[2
1
]
Y.
Yo
o
,
“
Hy
p
e
rp
a
ra
m
e
ter
o
p
ti
m
i
z
a
ti
o
n
o
f
d
e
e
p
n
e
u
ra
l
n
e
tw
o
rk
u
sin
g
u
n
i
v
a
riate
d
y
n
a
m
ic
e
n
c
o
d
in
g
a
lg
o
rit
h
m
fo
r
se
a
rc
h
e
s,”
Kn
o
wled
g
e
-
Ba
se
d
S
y
st
e
ms
,
v
o
l.
1
7
8
,
p
p
.
7
4
–
8
3
,
2
0
1
9
,
d
o
i:
1
0
.
1
0
1
6
/
j.
k
n
o
s
y
s.2
0
1
9
.
0
4
.
0
1
9
.
[2
2
]
M
.
Zh
a
n
g
,
H.
L
i,
J.
L
y
u
,
S
.
H.
Li
n
g
,
a
n
d
S
.
S
tev
e
n
,
“
M
u
lt
i
-
lev
e
l
C
NN
fo
r
lu
n
g
n
o
d
u
le
c
las
sifica
ti
o
n
with
G
a
u
ss
ian
P
ro
c
e
ss
a
ss
isted
h
y
p
e
rp
a
ra
m
e
ter
o
p
ti
m
iza
ti
o
n
,
”
0
2
-
Ja
n
-
2
0
1
9
.
[On
l
in
e
].
Av
a
il
a
b
le:
h
tt
p
:
//
a
rx
i
v
.
o
r
g
/a
b
s/1
9
0
1
.
0
0
2
7
6
.
[Ac
c
e
ss
e
d
:
1
2
-
Oc
t
-
2
0
2
1
].
[2
3
]
S.
H
.
G
u
p
t
a
,
A
.
S
h
a
r
m
a
,
M
.
M
o
h
t
a
,
a
n
d
A
.
R
a
j
a
w
a
t
,
“
H
a
n
d
m
o
v
e
m
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
f
r
o
m
m
e
a
s
u
r
e
d
s
c
a
t
t
e
r
i
n
g
p
a
r
a
m
e
t
e
r
s
u
s
i
n
g
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
M
e
a
s
u
r
e
m
e
n
t
,
v
o
l
.
1
5
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
a
s
u
r
e
m
e
n
t
.
2
0
1
9
.
1
0
7
2
5
8
.
[2
4
]
W.
Ku
su
m
a
,
Z.
S
a
ri,
A.
M
in
a
rn
o
,
H.
Wi
b
o
w
o
,
a
n
d
D.
Ak
b
i
,
“
P
e
rfo
r
m
a
n
c
e
Co
m
p
a
risso
n
Hu
m
a
n
Ac
ti
v
it
y
Re
c
o
g
n
it
io
n
u
sin
g
S
imp
le
Li
n
e
a
r
M
e
t
h
o
d
,
”
KI
NET
IK
,
v
o
l.
5
,
n
o
.
1
,
p
p
.
1
–
1
0
,
2
0
2
0
,
d
o
i:
1
0
.
2
2
2
1
9
/k
in
e
ti
k
.
v
5
i
1
.
1
0
2
5
.
[2
5
]
A.
G
a
lán
-
M
e
rc
a
n
t,
A.
Ortiz,
E.
H
e
rre
ra
-
Vie
d
m
a
,
M
.
To
m
a
s,
B.
F
e
r
n
a
n
d
e
s,
a
n
d
J.
M
o
ra
;
-
M
u
n
o
z
,
“
As
se
ss
in
g
p
h
y
sic
a
l
a
c
ti
v
it
y
a
n
d
f
u
n
c
ti
o
n
a
l
fit
n
e
ss
lev
e
l
u
si
n
g
c
o
n
v
o
lu
t
io
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s,”
K
n
o
w
led
g
e
-
B
a
se
d
S
y
ste
ms
,
v
o
l.
1
8
5
,
2
0
1
9
,
d
o
i:
1
0
.
1
0
1
6
/j
.
k
n
o
sy
s.
2
0
1
9
.
1
0
4
9
3
9
.
[2
6
]
K.
Ch
e
n
,
D.
Z
h
a
n
g
,
L.
Ya
o
,
B.
G
u
o
,
Z
.
Yu
,
a
n
d
Y.
Li
u
,
“
De
e
p
lea
rn
in
g
fo
r
se
n
so
r
-
b
a
se
d
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
io
n
:
Ov
e
rv
i
e
w,
c
h
a
ll
e
n
g
e
s,
a
n
d
o
p
p
o
rt
u
n
it
ies
,
”
AC
M
Co
mp
u
ti
n
g
S
u
rv
e
y
s
,
v
o
l.
5
4
,
n
o
.
4
,
J
u
l.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
4
5
/
3
4
4
7
7
4
4
.
[2
7
]
Z.
Kh
a
n
a
n
d
J.
Ah
m
a
d
,
“
Atten
t
io
n
in
d
u
c
e
d
m
u
l
ti
-
h
e
a
d
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
f
o
r
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
i
ti
o
n
,
”
Ap
p
li
e
d
S
o
ft
Co
mp
u
ti
n
g
,
v
o
l.
1
1
0
,
2
0
2
1
,
d
o
i:
1
0
.
1
0
1
6
/j
.
a
so
c
.
2
0
2
1
.
1
0
7
6
7
1
.
[2
8
]
S
.
K.
Ch
a
ll
a
,
A
.
Ku
m
a
r,
a
n
d
V.
B.
S
e
m
wa
l,
“
A
m
u
lt
i
b
ra
n
c
h
CN
N
-
BiL
S
TM
m
o
d
e
l
f
o
r
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
it
i
o
n
u
sin
g
we
a
ra
b
le se
n
so
r
d
a
ta,”
T
h
e
Vi
su
a
l
Co
mp
u
ter
,
p
p
.
1
–
1
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/s
0
0
3
7
1
-
0
2
1
-
0
2
2
8
3
-
3.
[2
9
]
S
.
M
e
k
ru
k
sa
v
a
n
ich
a
n
d
A.
Jit
p
a
tt
a
n
a
k
u
l,
“
Bio
m
e
tri
c
u
se
r
i
d
e
n
ti
fic
a
ti
o
n
b
a
se
d
o
n
h
u
m
a
n
a
c
ti
v
it
y
re
c
o
g
n
i
ti
o
n
u
sin
g
we
a
ra
b
le
se
n
so
rs:
An
e
x
p
e
rime
n
t
u
si
n
g
d
e
e
p
lea
rn
i
n
g
m
o
d
e
ls,”
E
lec
tro
n
ics
,
v
o
l.
1
0
,
n
o
.
3
,
2
0
2
1
,
d
o
i:
1
0
.
3
3
9
0
/ele
c
tro
n
ics
1
0
0
3
0
3
0
8
.
[3
0
]
Z.
F
u
,
X.
He
,
E.
Wan
g
,
J.
H
u
o
,
J.
Hu
a
n
g
,
a
n
d
D.
W
u
,
“
P
e
rso
n
a
li
z
e
d
Hu
m
a
n
Ac
ti
v
it
y
Re
c
o
g
n
it
i
o
n
Ba
se
d
o
n
In
teg
ra
te
d
Wea
ra
b
le S
e
n
so
r
a
n
d
Tran
sfe
r
Le
a
rn
in
g
,
”
S
e
n
s
o
rs
,
v
o
l.
2
1
,
n
o
.
3
,
2
0
2
1
,
d
o
i:
1
0
.
3
3
9
0
/s2
1
0
3
0
8
8
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.