<s>
3D	B-Algorithm
reconstruction	I-Algorithm
from	I-Algorithm
multiple	I-Algorithm
images	I-Algorithm
is	O
the	O
creation	O
of	O
three-dimensional	O
models	O
from	O
a	O
set	O
of	O
images	O
.	O
</s>
<s>
This	O
process	O
is	O
referred	O
to	O
as	O
triangulation	B-Algorithm
.	O
</s>
<s>
In	O
recent	O
decades	O
,	O
there	O
is	O
an	O
important	O
demand	O
for	O
3D	O
content	O
for	O
computer	O
graphics	O
,	O
virtual	B-Application
reality	I-Application
and	O
communication	O
,	O
triggering	O
a	O
change	O
in	O
emphasis	O
for	O
the	O
requirements	O
.	O
</s>
<s>
Camera	B-Algorithm
calibration	I-Algorithm
consists	O
of	O
intrinsic	O
and	O
extrinsic	O
parameters	O
,	O
without	O
which	O
at	O
some	O
level	O
no	O
arrangement	O
of	O
algorithms	O
can	O
work	O
.	O
</s>
<s>
The	O
dotted	O
line	O
between	O
Calibration	O
and	O
Depth	O
determination	O
represents	O
that	O
the	O
camera	B-Algorithm
calibration	I-Algorithm
is	O
usually	O
required	O
for	O
determining	O
depth	O
.	O
</s>
<s>
The	O
correspondence	B-Algorithm
problem	I-Algorithm
,	O
finding	O
matches	O
between	O
two	O
images	O
so	O
the	O
position	O
of	O
the	O
matched	O
elements	O
can	O
then	O
be	O
triangulated	O
in	O
3D	O
space	O
is	O
the	O
key	O
issue	O
here	O
.	O
</s>
<s>
Once	O
you	O
have	O
the	O
multiple	O
depth	O
maps	O
you	O
have	O
to	O
combine	O
them	O
to	O
create	O
a	O
final	O
mesh	O
by	O
calculating	O
depth	O
and	O
projecting	O
out	O
of	O
the	O
camera	O
–	O
registration	B-Algorithm
.	O
</s>
<s>
Camera	B-Algorithm
calibration	I-Algorithm
will	O
be	O
used	O
to	O
identify	O
where	O
the	O
many	O
meshes	O
created	O
by	O
depth	O
maps	O
can	O
be	O
combined	O
to	O
develop	O
a	O
larger	O
one	O
,	O
providing	O
more	O
than	O
one	O
view	O
for	O
observation	O
.	O
</s>
<s>
Design	O
a	O
cost	O
function	O
,	O
which	O
considers	O
the	O
intrinsic	O
parameters	O
as	O
arguments	O
and	O
the	O
fundamental	B-General_Concept
matrices	I-General_Concept
as	O
parameters	O
.	O
</s>
<s>
is	O
defined	O
as	O
the	O
fundamental	B-General_Concept
matrix	I-General_Concept
,	O
and	O
as	O
intrinsic	O
parameters	O
matrices	O
.	O
</s>
<s>
The	O
concept	O
of	O
stratification	O
is	O
closely	O
related	O
to	O
the	O
series	O
of	O
transformations	O
on	O
geometric	O
entities	O
:	O
in	O
the	O
projective	O
stratum	O
is	O
a	O
series	O
of	O
projective	O
transformations	O
(	O
a	O
homography	B-Algorithm
)	O
,	O
in	O
the	O
affine	O
stratum	O
is	O
a	O
series	O
of	O
affine	B-Algorithm
transformations	I-Algorithm
,	O
and	O
in	O
Euclidean	O
stratum	O
is	O
a	O
series	O
of	O
Euclidean	O
transformations	O
.	O
</s>
<s>
To	O
reduce	O
the	O
effect	O
of	O
noise	O
,	O
we	O
usually	O
use	O
more	O
equations	O
than	O
necessary	O
and	O
solve	O
with	O
least	B-Algorithm
squares	I-Algorithm
.	O
</s>
<s>
For	O
example	O
,	O
in	O
a	O
typical	O
null-space	O
problem	O
formulation	O
Ax	O
=	O
0	O
(	O
like	O
the	O
DLT	O
algorithm	O
)	O
,	O
the	O
square	O
of	O
the	O
residual	O
||Ax||	O
is	O
being	O
minimized	O
with	O
the	O
least	B-Algorithm
squares	I-Algorithm
method	I-Algorithm
.	O
</s>
<s>
Medical	B-Application
imaging	I-Application
techniques	I-Application
like	O
CT	O
scanning	O
and	O
MRI	O
are	O
expensive	O
,	O
and	O
although	O
CT	O
scans	O
are	O
accurate	O
,	O
they	O
can	O
induce	O
high	O
radiation	O
doses	O
which	O
is	O
a	O
risk	O
for	O
patients	O
with	O
certain	O
diseases	O
.	O
</s>
<s>
The	O
first	O
step	O
is	O
to	O
extract	O
the	O
corresponding	B-Algorithm
points	I-Algorithm
in	O
two	O
x-ray	O
images	O
.	O
</s>
<s>
The	O
reconstruction	O
is	O
only	O
possible	O
where	O
there	O
are	O
Stereo	O
Corresponding	B-Algorithm
Points	I-Algorithm
(	O
SCPs	O
)	O
.	O
</s>
