<s>
Volumetric	B-Algorithm
capture	I-Algorithm
or	O
volumetric	B-Algorithm
video	I-Algorithm
is	O
a	O
technique	O
that	O
captures	O
a	O
three-dimensional	O
space	O
,	O
such	O
as	O
a	O
location	O
or	O
performance	O
.	O
</s>
<s>
This	O
type	O
of	O
volumography	O
acquires	O
data	O
that	O
can	O
be	O
viewed	O
on	O
flat	O
screens	O
as	O
well	O
as	O
using	O
3D	B-Algorithm
displays	I-Algorithm
and	O
VR	B-Device
goggles	I-Device
.	O
</s>
<s>
Consumer-facing	O
formats	O
are	O
numerous	O
and	O
the	O
required	O
motion	B-Application
capture	I-Application
techniques	O
lean	O
on	O
computer	O
graphics	O
,	O
photogrammetry	B-Application
,	O
and	O
other	O
computation-based	O
methods	O
.	O
</s>
<s>
The	O
viewer	O
generally	O
experiences	O
the	O
result	O
in	O
a	O
real-time	B-General_Concept
engine	O
and	O
has	O
direct	O
input	O
in	O
exploring	O
the	O
generated	O
volume	O
.	O
</s>
<s>
Holograms	B-Algorithm
and	O
3D	O
real-world	O
visuals	O
have	O
featured	O
prominently	O
in	O
Star	B-Application
Wars	I-Application
,	O
Blade	B-Application
Runner	I-Application
,	O
and	O
many	O
other	O
science-fiction	O
productions	O
over	O
the	O
years	O
.	O
</s>
<s>
Volumetric	B-Algorithm
video	I-Algorithm
is	O
the	O
logical	O
next	O
step	O
after	O
stereoscopic	O
movies	O
and	O
360°	O
videos	O
in	O
that	O
it	O
combines	O
the	O
visual	O
quality	O
of	O
photography	O
with	O
the	O
immersion	O
and	O
interactivity	O
of	O
spatialized	O
content	O
and	O
could	O
prove	O
to	O
be	O
the	O
most	O
important	O
development	O
in	O
the	O
recording	O
of	O
human	O
performance	O
since	O
the	O
creation	O
of	O
contemporary	O
cinema	O
.	O
</s>
<s>
The	O
first	O
volumetric	B-Algorithm
video	I-Algorithm
market	O
report	O
came	O
out	O
in	O
May	O
2018	O
and	O
the	O
first	O
online	O
publication	O
dedicated	O
to	O
volumetric	B-Algorithm
video	I-Algorithm
was	O
released	O
in	O
November	O
2019	O
-	O
""	O
.	O
</s>
<s>
Visual	O
effects	O
in	O
movies	O
and	O
video	O
games	O
paved	O
the	O
way	O
for	O
advances	O
in	O
photogrammetry	B-Application
,	O
scanning	O
devices	O
,	O
and	O
the	O
computational	O
backend	O
to	O
handle	O
the	O
data	O
received	O
from	O
these	O
new	O
intensive	O
methods	O
.	O
</s>
<s>
LIDAR	O
scanning	O
describes	O
a	O
survey	O
method	O
that	O
uses	O
laser-sampled	O
points	O
densely	O
packed	O
to	O
scan	O
static	O
objects	O
into	O
a	O
point	B-Algorithm
cloud	I-Algorithm
.	O
</s>
<s>
This	O
requires	O
physical	O
scanners	B-Algorithm
and	O
produces	O
enormous	O
amounts	O
of	O
data	O
.	O
</s>
<s>
In	O
2007	O
the	O
band	O
Radiohead	O
used	O
it	O
extensively	O
to	O
create	O
a	O
music	O
video	O
for	O
"	O
House	O
of	O
Cards	O
"	O
,	O
capturing	O
point	B-Algorithm
cloud	I-Algorithm
performances	O
of	O
the	O
singer	O
's	O
face	O
and	O
of	O
select	O
environments	O
in	O
one	O
of	O
the	O
first	O
uses	O
of	O
this	O
technology	O
for	O
volumetric	B-Algorithm
capture	I-Algorithm
.	O
</s>
<s>
Director	O
James	O
Frost	O
collaborated	O
with	O
media	O
artist	O
Aaron	O
Koblin	O
to	O
capture	O
3D	O
point-clouds	B-Algorithm
used	O
for	O
this	O
music	O
clip	O
,	O
and	O
while	O
the	O
final	O
output	O
of	O
this	O
work	O
was	O
still	O
a	O
rendered	O
flat	O
representation	O
of	O
the	O
data	O
,	O
the	O
capture	O
and	O
mindset	O
of	O
the	O
authors	O
was	O
already	O
ahead	O
of	O
its	O
time	O
.	O
</s>
<s>
Point	B-Algorithm
clouds	I-Algorithm
,	O
being	O
distinct	O
samples	O
of	O
three-dimensional	O
space	O
with	O
position	O
and	O
color	O
,	O
create	O
a	O
high	O
fidelity	O
representation	O
of	O
the	O
real	O
world	O
with	O
a	O
huge	O
amount	O
of	O
data	O
.	O
</s>
<s>
However	O
,	O
viewing	O
this	O
data	O
in	O
real-time	B-General_Concept
was	O
not	O
yet	O
possible	O
.	O
</s>
<s>
In	O
2010	O
Microsoft	O
brought	O
the	O
Kinect	B-Algorithm
to	O
the	O
market	O
,	O
a	O
consumer	O
product	O
that	O
used	O
structured	O
light	O
in	O
the	O
infrared	O
spectrum	O
to	O
generate	O
a	O
3D	O
mesh	O
from	O
its	O
camera	O
.	O
</s>
<s>
While	O
the	O
intent	O
was	O
to	O
facilitate	O
and	O
innovate	O
in	O
user	O
input	O
and	O
gameplay	O
,	O
it	O
was	O
very	O
quickly	O
adapted	O
as	O
a	O
generic	O
capture	O
device	O
for	O
3D	O
data	O
in	O
the	O
volumetric	B-Algorithm
capture	I-Algorithm
community	O
.	O
</s>
<s>
Artists	O
and	O
hobbyists	O
started	O
to	O
make	O
tools	O
and	O
projects	O
around	O
the	O
affordable	O
device	O
,	O
sparking	O
a	O
growing	O
interest	O
in	O
volumetric	B-Algorithm
capture	I-Algorithm
as	O
a	O
creative	O
medium	O
.	O
</s>
<s>
Researchers	O
at	O
Microsoft	O
then	O
constructed	O
an	O
entire	O
capture	O
stage	O
using	O
multiple	O
cameras	O
,	O
Kinect	B-Algorithm
devices	O
,	O
and	O
algorithms	O
that	O
generated	O
a	O
full	O
volumetric	B-Algorithm
capture	I-Algorithm
from	O
the	O
combined	O
optical	O
and	O
depth	O
information	O
.	O
</s>
<s>
This	O
is	O
now	O
the	O
,	O
used	O
today	O
as	O
part	O
of	O
both	O
their	O
research	O
division	O
and	O
in	O
certain	O
select	O
commercial	O
experiences	O
such	O
as	O
the	O
Blade	B-Application
Runner	I-Application
2049	O
VR	O
experience	O
.	O
</s>
<s>
While	O
this	O
remains	O
a	O
very	O
interesting	O
setup	O
for	O
the	O
high-end	O
market	O
,	O
the	O
affordable	O
price	O
of	O
a	O
single	O
Kinect	B-Algorithm
device	O
led	O
more	O
experimental	O
artists	O
and	O
independent	O
directors	O
to	O
become	O
active	O
in	O
the	O
volumetric	B-Algorithm
capture	I-Algorithm
field	O
.	O
</s>
<s>
EF	O
EVE™	O
supports	O
unlimited	O
number	O
of	O
Azure	O
Kinect	B-Algorithm
sensors	O
on	O
one	O
PC	O
giving	O
full	O
volumetric	B-Algorithm
capture	I-Algorithm
with	O
easy	O
setup	O
.	O
</s>
<s>
Depthkit	O
is	O
a	O
software	O
suite	O
that	O
allows	O
the	O
capture	O
of	O
geometry	O
data	O
with	O
one	O
structured	O
light	O
sensor	O
including	O
the	O
Azure	O
Kinect	B-Algorithm
,	O
as	O
well	O
as	O
high	O
quality	O
color	O
detail	O
from	O
an	O
attached	O
witness	O
camera	O
.	O
</s>
<s>
Photogrammetry	B-Application
describes	O
the	O
process	O
of	O
measuring	O
data	O
based	O
on	O
photographic	O
reference	O
.	O
</s>
<s>
While	O
being	O
as	O
old	O
as	O
photography	O
itself	O
,	O
only	O
through	O
advances	O
over	O
the	O
years	O
in	O
volumetric	B-Algorithm
capture	I-Algorithm
research	O
has	O
it	O
now	O
become	O
possible	O
to	O
capture	O
more	O
and	O
more	O
geometry	O
and	O
texture	O
detail	O
from	O
a	O
large	O
number	O
of	O
input	O
images	O
.	O
</s>
<s>
The	O
result	O
is	O
usually	O
split	O
into	O
two	O
composited	O
sources	O
,	O
static	O
geometry	O
and	O
full	O
performance	B-Application
capture	I-Application
.	O
</s>
<s>
For	O
static	O
geometry	O
,	O
sets	O
that	O
are	O
captured	O
with	O
a	O
large	O
number	O
of	O
overlapping	O
digital	O
images	O
are	O
then	O
aligned	O
to	O
each	O
other	O
using	O
similar	O
features	O
in	O
the	O
images	O
and	O
used	O
as	O
a	O
base	O
for	O
triangulation	B-Algorithm
and	O
depth	O
estimation	O
.	O
</s>
<s>
Full	O
performance	B-Application
capture	I-Application
,	O
however	O
,	O
uses	O
an	O
array	O
of	O
video	O
cameras	O
to	O
capture	O
real-time	B-General_Concept
information	O
.	O
</s>
<s>
Those	O
synchronized	O
cameras	O
are	O
then	O
used	O
frame-by-frame	O
to	O
generate	O
a	O
set	O
of	O
points	O
or	O
geometry	O
that	O
can	O
be	O
played	O
back	O
at	O
speed	O
,	O
resulting	O
in	O
the	O
full	O
volumetric	B-Algorithm
performance	I-Algorithm
capture	I-Algorithm
that	O
can	O
be	O
composited	O
into	O
any	O
environment	O
.	O
</s>
<s>
In	O
2008	O
,	O
4DViews	O
installed	O
a	O
first	O
volumetric	B-Algorithm
video	I-Algorithm
capture	O
system	O
at	O
DigiCast	O
studio	O
in	O
Tokyo	O
(	O
JP	O
)	O
.	O
</s>
<s>
Later	O
in	O
2015	O
,	O
8i	O
contributed	O
in	O
the	O
field	O
,	O
and	O
recently	O
Intel	O
,	O
Microsoft	O
,	O
and	O
Samsung	O
have	O
joined	O
in	O
by	O
creating	O
their	O
own	O
capture	O
stages	O
for	O
performance	B-Application
capture	I-Application
and	O
photogrammetry	B-Application
.	O
</s>
<s>
As	O
volumetric	B-Algorithm
video	I-Algorithm
developed	O
into	O
a	O
commercially	O
applicable	O
approach	O
to	O
environment	O
and	O
performance	B-Application
capture	I-Application
,	O
the	O
ability	O
to	O
move	O
about	O
the	O
results	O
with	O
six	O
degrees	O
of	O
freedom	O
and	O
true	O
stereoscopy	O
necessitated	O
a	O
new	O
type	O
of	O
display	O
device	O
.	O
</s>
<s>
With	O
the	O
rise	O
of	O
consumer-facing	O
VR	O
in	O
2016	O
through	O
devices	O
such	O
as	O
the	O
Oculus	B-Device
Rift	I-Device
and	O
HTC	O
Vive	O
,	O
this	O
was	O
suddenly	O
possible	O
.	O
</s>
<s>
With	O
the	O
rise	O
of	O
360°	O
video	O
content	O
,	O
the	O
demand	O
for	O
6-DOF	O
capture	O
is	O
rising	O
,	O
and	O
VR	O
in	O
particular	O
drives	O
the	O
applications	O
for	O
this	O
technology	O
,	O
slowly	O
fusing	O
cinema	O
,	O
games	O
and	O
art	O
with	O
the	O
field	O
of	O
volumetric	B-Algorithm
capture	I-Algorithm
research	O
.	O
</s>
<s>
Facebook	B-Application
is	O
using	O
this	O
idea	O
in	O
its	O
Surround360	O
camera	O
family	O
to	O
capture	O
360°	O
video	O
footage	O
that	O
is	O
getting	O
stitched	O
with	O
the	O
help	O
of	O
distance	O
maps	O
.	O
</s>
<s>
Different	O
workflows	O
to	O
generate	O
volumetric	B-Algorithm
video	I-Algorithm
are	O
currently	O
available	O
.	O
</s>
<s>
This	O
approach	O
generates	O
a	O
more	O
traditional	O
3D	O
triangle	B-Algorithm
mesh	I-Algorithm
similar	O
to	O
the	O
geometry	O
used	O
for	O
computer	O
games	O
and	O
visual	O
effects	O
.	O
</s>
<s>
Photogrammetry	B-Application
is	O
usually	O
used	O
as	O
a	O
base	O
for	O
static	O
meshes	O
,	O
and	O
is	O
then	O
augmented	O
with	O
performance	B-Application
capture	I-Application
of	O
talent	O
via	O
the	O
same	O
underlying	O
technology	O
of	O
videogrammetry	B-Algorithm
.	O
</s>
<s>
The	O
playback	O
is	O
usually	O
handled	O
by	O
a	O
real-time	B-General_Concept
engine	O
and	O
resembles	O
a	O
traditional	O
game	O
pipeline	O
in	O
implementation	O
,	O
allowing	O
interactive	O
lighting	O
changes	O
and	O
creative	O
and	O
archivable	O
ways	O
of	O
compositing	O
static	O
and	O
animated	O
meshes	O
together	O
.	O
</s>
<s>
Recently	O
the	O
spotlight	O
has	O
shifted	O
towards	O
point-based	B-Algorithm
volumetric	B-Algorithm
capture	I-Algorithm
.	O
</s>
<s>
Points	O
can	O
either	O
be	O
scattered	O
on	O
triangle	B-Algorithm
meshes	I-Algorithm
with	O
pre-computed	O
lighting	O
,	O
or	O
used	O
directly	O
from	O
a	O
LIDAR	O
scanner	O
.	O
</s>
<s>
The	O
final	O
product	O
can	O
then	O
be	O
viewed	O
either	O
as	O
a	O
flat	O
rendering	O
of	O
the	O
captured	O
data	O
,	O
or	O
interactively	O
in	O
a	O
VR	B-Device
headset	I-Device
.	O
</s>
<s>
While	O
one	O
goal	O
,	O
with	O
the	O
point-based	B-Algorithm
approach	O
to	O
volumetric	B-Algorithm
capture	I-Algorithm
,	O
is	O
to	O
stream	O
point	O
data	O
from	O
cloud	O
to	O
the	O
user	O
at	O
home	O
,	O
allowing	O
the	O
creating	O
and	O
dissemination	O
of	O
realistic	O
virtual	O
worlds	O
on	O
demand	O
-	O
a	O
second	O
goal	O
more	O
recently	O
considered	O
would	O
be	O
a	O
real-time	B-General_Concept
data	O
stream	O
of	O
live	O
events	O
.	O
</s>
<s>
As	O
volumetric	B-Algorithm
video	I-Algorithm
evolves	O
into	O
global	O
capture	O
and	O
the	O
display	O
hardware	O
evolves	O
to	O
match	O
,	O
we	O
will	O
enter	O
into	O
an	O
era	O
of	O
true	O
immersion	O
where	O
the	O
nuances	O
of	O
captured	O
environment	O
combined	O
with	O
those	O
of	O
captured	O
performances	O
will	O
convey	O
emotionality	O
in	O
a	O
whole	O
new	O
medium	O
,	O
blurring	O
the	O
boundaries	O
between	O
real	O
and	O
virtual	O
worlds	O
.	O
</s>
<s>
One	O
area	O
of	O
concern	O
with	O
the	O
growing	O
field	O
of	O
volumetric	B-Algorithm
capture	I-Algorithm
is	O
the	O
shrinking	O
of	O
demand	O
for	O
traditional	O
skillsets	O
like	O
modeling	O
,	O
lighting	O
,	O
animation	O
,	O
etc	O
.	O
</s>
<s>
However	O
,	O
while	O
in	O
future	O
the	O
stack	O
of	O
production-oriented	O
volumetric	B-Algorithm
capture	I-Algorithm
technologies	O
will	O
grow	O
and	O
grow	O
,	O
so	O
too	O
will	O
the	O
demand	O
for	O
traditional	O
skillsets	O
.	O
</s>
<s>
Volumetric	B-Algorithm
capture	I-Algorithm
excels	O
at	O
capturing	O
static	O
data	O
or	O
pre-rendered	O
animated	O
footage	O
.	O
</s>
<s>
By	O
combining	O
these	O
real-life	O
set	O
capture	O
with	O
the	O
volumetric	B-Algorithm
captures	I-Algorithm
of	O
additional	O
CG	O
elements	O
,	O
we	O
will	O
be	O
able	O
to	O
blend	O
real-life	O
and	O
our	O
imagination	O
in	O
a	O
way	O
that	O
we	O
have	O
only	O
previously	O
been	O
able	O
to	O
do	O
on	O
a	O
flat-screen	O
,	O
creating	O
new	O
fields	O
in	O
the	O
areas	O
like	O
compositing	O
and	O
VFX	O
.	O
</s>
<s>
As	O
every	O
medium	O
creates	O
its	O
own	O
visual	O
language	O
,	O
rules	O
and	O
creative	O
approaches	O
,	O
volumetric	B-Algorithm
video	I-Algorithm
is	O
still	O
at	O
the	O
infancy	O
of	O
this	O
.	O
</s>
<s>
There	O
is	O
no	O
standard	O
yet	O
that	O
generated	O
volumetric	B-Algorithm
video	I-Algorithm
and	O
makes	O
it	O
experienceable	O
at	O
home	O
.	O
</s>
