<s>
Subjective	B-Device
video	I-Device
quality	I-Device
is	O
video	B-Device
quality	I-Device
as	O
experienced	O
by	O
humans	O
.	O
</s>
<s>
Measuring	O
subjective	B-Device
video	I-Device
quality	I-Device
is	O
necessary	O
because	O
objective	O
quality	O
assessment	O
algorithms	O
such	O
as	O
PSNR	O
have	O
been	O
shown	O
to	O
correlate	O
poorly	O
with	O
subjective	O
ratings	O
.	O
</s>
<s>
Subjective	B-Device
video	I-Device
quality	I-Device
tests	O
are	O
psychophysical	O
experiments	O
in	O
which	O
a	O
number	O
of	O
viewers	O
rate	O
a	O
given	O
set	O
of	O
stimuli	O
.	O
</s>
<s>
In	O
subjective	B-Device
video	I-Device
quality	I-Device
tests	O
,	O
typically	O
,	O
SRCs	O
(	O
"	O
Sources	O
"	O
,	O
i.e.	O
</s>
<s>
The	O
main	O
idea	O
of	O
measuring	O
subjective	B-Device
video	I-Device
quality	I-Device
is	O
similar	O
to	O
the	O
mean	O
opinion	O
score	O
(	O
MOS	O
)	O
evaluation	O
for	O
audio	O
.	O
</s>
<s>
To	O
evaluate	O
the	O
subjective	B-Device
video	I-Device
quality	I-Device
of	O
a	O
video	O
processing	O
system	O
,	O
the	O
following	O
steps	O
are	O
typically	O
taken	O
:	O
</s>
<s>
There	O
should	O
be	O
no	O
visible	O
coding	B-Algorithm
artifacts	I-Algorithm
or	O
other	O
properties	O
that	O
would	O
lower	O
the	O
quality	O
of	O
the	O
original	O
sequence	O
.	O
</s>
<s>
For	O
example	O
,	O
to	O
test	O
the	O
quality	O
of	O
a	O
video	B-Application
codec	I-Application
,	O
independent	O
variables	O
may	O
be	O
the	O
video	O
encoding	O
software	O
,	O
a	O
target	O
bitrate	O
,	O
and	O
the	O
target	O
resolution	O
of	O
the	O
processed	O
sequence	O
.	O
</s>
<s>
The	O
minimum	O
number	O
of	O
subjects	O
that	O
are	O
required	O
for	O
a	O
subjective	B-Device
video	I-Device
quality	I-Device
study	O
is	O
not	O
strictly	O
defined	O
.	O
</s>
<s>
However	O
,	O
most	O
recommendations	O
for	O
the	O
number	O
of	O
subjects	O
have	O
been	O
designed	O
for	O
measuring	O
video	B-Device
quality	I-Device
encountered	O
by	O
a	O
home	O
television	O
or	O
PC	O
user	O
,	O
where	O
the	O
range	O
and	O
diversity	O
of	O
distortions	O
tend	O
to	O
be	O
limited	O
(	O
e.g.	O
,	O
to	O
encoding	O
artifacts	B-Algorithm
only	O
)	O
.	O
</s>
<s>
There	O
is	O
an	O
ongoing	O
discussion	O
in	O
the	O
QoE	O
community	O
as	O
to	O
whether	O
a	O
viewer	O
's	O
cultural	O
,	O
social	O
,	O
or	O
economic	O
background	O
has	O
a	O
significant	O
impact	O
on	O
the	O
obtained	O
subjective	B-Device
video	I-Device
quality	I-Device
results	O
.	O
</s>
<s>
A	O
systematic	O
study	O
involving	O
six	O
laboratories	O
in	O
four	O
countries	O
found	O
no	O
statistically	O
significant	O
impact	O
of	O
subject	O
's	O
language	O
and	O
culture	O
/	O
country	O
of	O
origin	O
on	O
video	B-Device
quality	I-Device
ratings	O
.	O
</s>
<s>
Crowdsourcing	O
has	O
recently	O
been	O
used	O
for	O
subjective	B-Device
video	I-Device
quality	I-Device
evaluation	O
,	O
and	O
more	O
generally	O
,	O
in	O
the	O
context	O
of	O
Quality	O
of	O
Experience	O
.	O
</s>
<s>
A	O
number	O
of	O
subjective	O
picture	O
and	O
video	B-Device
quality	I-Device
databases	O
based	O
on	O
such	O
studies	O
have	O
been	O
made	O
publicly	O
available	O
by	O
research	O
institutes	O
.	O
</s>
