<s>
In	O
the	O
field	O
of	O
computer	B-Application
vision	I-Application
,	O
any	O
two	O
images	O
of	O
the	O
same	O
planar	O
surface	O
in	O
space	O
are	O
related	O
by	O
a	O
homography	B-Algorithm
(	O
assuming	O
a	O
pinhole	B-Algorithm
camera	I-Algorithm
model	I-Algorithm
)	O
.	O
</s>
<s>
This	O
has	O
many	O
practical	O
applications	O
,	O
such	O
as	O
image	B-Algorithm
rectification	I-Algorithm
,	O
image	B-Algorithm
registration	I-Algorithm
,	O
or	O
camera	O
motion	O
—	O
rotation	O
and	O
translation	O
—	O
between	O
two	O
images	O
.	O
</s>
<s>
Once	O
camera	B-Algorithm
resectioning	I-Algorithm
has	O
been	O
done	O
from	O
an	O
estimated	O
homography	B-Algorithm
matrix	O
,	O
this	O
information	O
may	O
be	O
used	O
for	O
navigation	O
,	O
or	O
to	O
insert	O
models	O
of	O
3D	O
objects	O
into	O
an	O
image	O
or	O
video	O
,	O
so	O
that	O
they	O
are	O
rendered	O
with	O
the	O
correct	O
perspective	O
and	O
appear	O
to	O
have	O
been	O
part	O
of	O
the	O
original	O
scene	O
(	O
see	O
Augmented	B-General_Concept
reality	I-General_Concept
)	O
.	O
</s>
<s>
is	O
the	O
rotation	B-Algorithm
matrix	I-Algorithm
by	O
which	O
b	O
is	O
rotated	O
in	O
relation	O
to	O
a	O
;	O
t	O
is	O
the	O
translation	O
vector	O
from	O
a	O
to	O
b	O
;	O
n	O
and	O
d	O
are	O
the	O
normal	O
vector	O
of	O
the	O
plane	O
and	O
the	O
distance	O
from	O
origin	O
to	O
the	O
plane	O
respectively	O
.	O
</s>
<s>
When	O
the	O
image	O
region	O
in	O
which	O
the	O
homography	B-Algorithm
is	O
computed	O
is	O
small	O
or	O
the	O
image	O
has	O
been	O
acquired	O
with	O
a	O
large	O
focal	B-Algorithm
length	I-Algorithm
,	O
an	O
affine	O
homography	B-Algorithm
is	O
a	O
more	O
appropriate	O
model	O
of	O
image	O
displacements	O
.	O
</s>
