<s>
Audio	O
Video	O
Coding	O
Standard	O
(	O
AVS	B-Algorithm
)	O
refers	O
to	O
the	O
digital	O
audio	O
and	O
digital	O
video	O
series	O
compression	O
standard	O
formulated	O
by	O
the	O
Audio	O
and	O
Video	O
coding	O
standard	O
workgroup	O
of	O
China	O
.	O
</s>
<s>
The	O
first	O
generation	O
AVS	B-Algorithm
standard	O
includes	O
“	O
Information	O
Technology	O
,	O
Advanced	O
Audio	O
Video	O
Coding	O
,	O
Part	B-Algorithm
2	I-Algorithm
:	O
Video	O
”	O
(	O
AVS1	B-Algorithm
)	O
and	O
“	O
Information	O
Technology	O
,	O
Advanced	O
Audio	O
Video	O
Coding	O
Part	O
16	O
:	O
Radio	O
Television	O
Video	O
”	O
(	O
AVS+	O
)	O
.	O
</s>
<s>
For	O
the	O
second	O
generation	O
,	O
referred	O
to	O
as	O
AVS2	B-Algorithm
,	O
the	O
primary	O
application	O
target	O
was	O
Ultra-high-definition	B-Device
television	I-Device
video	O
,	O
supporting	O
the	O
efficient	O
compression	O
of	O
ultra-high-resolution	O
(	O
4K	B-Architecture
resolution	I-Architecture
and	O
above	O
)	O
,	O
high-dynamic-range	B-Application
videos	I-Application
,	O
and	O
was	O
published	O
as	O
IEEE	O
international	O
standard	O
IEEE	O
1857.4	O
.	O
</s>
<s>
An	O
industry	O
alliance	O
was	O
established	O
to	O
develop	O
and	O
promote	O
AVS	B-Algorithm
standards	O
.	O
</s>
<s>
AVS3	O
codec	B-General_Concept
added	O
to	O
DVB	B-Application
’s	O
media	O
delivery	O
toolbox	O
.	O
</s>
<s>
The	O
AVS	B-Algorithm
workgroup	O
was	O
founded	O
in	O
June	O
2002	O
to	O
cooperate	O
with	O
Chinese	O
enterprises	O
and	O
scientific	O
research	O
institutions	O
,	O
to	O
formulate	O
(	O
revise	O
)	O
common	O
technical	O
standards	O
such	O
as	O
digital	O
audio	O
and	O
digital	O
video	O
's	O
compression	O
,	O
decompression	O
,	O
processing	O
and	O
representation	O
,	O
thus	O
to	O
provide	O
efficient	O
and	O
economic	O
coding/decoding	O
technologies	O
for	O
digital	O
audio	O
and	O
digital	O
video	O
devices	O
and	O
systems	O
,	O
serving	O
the	O
high-resolution	O
digital	O
broadcasting	O
,	O
high-density	O
digital	O
laser	O
storage	O
media	O
,	O
wireless	O
broadband	O
multimedia	O
communication	O
,	O
Internet	O
broadband	O
streaming	O
media	O
and	O
other	O
applications	O
.	O
</s>
<s>
The	O
first	O
setback	O
was	O
when	O
China	O
did	O
not	O
use	O
AVS	B-Algorithm
for	O
its	O
own	O
digital	O
television	O
broadcast	O
system	O
in	O
2003	O
.	O
</s>
<s>
The	O
royalty	O
for	O
the	O
first	O
generation	O
AVS	B-Algorithm
standard	O
was	O
only	O
one	O
yuan	O
per	O
terminal	O
.	O
</s>
<s>
The	O
AVS	B-Algorithm
industry	O
alliance	O
is	O
the	O
abbreviation	O
for	O
Zhongguancun	O
audio	O
visual	O
industry	O
technology	O
innovation	O
alliance	O
.	O
</s>
<s>
In	O
May	O
2005	O
,	O
twelve	O
enterprises	O
(	O
units	O
)	O
of	O
TCL	O
Group	O
Co.	O
,	O
Ltd.	O
,	O
Skyworth	O
Group	O
Research	O
Institute	O
,	O
Huawei	O
Technology	O
Co.	O
,	O
Ltd.	O
,	O
Hisense	O
Group	O
Co.	O
,	O
Ltd.	O
,	O
Haier	O
Group	O
Co.	O
,	O
Ltd.	O
,	O
Beijing	O
Haier	O
Guangke	O
Co.	O
,	O
Ltd.	O
,	O
Inspur	O
Group	O
Co.	O
,	O
Ltd.	O
,	O
Joint	O
Source	O
Digital	O
Audio	O
Video	O
Technology	O
(	O
Beijing	O
)	O
Co.	O
,	O
Ltd.	O
,	O
New	O
Pudong	O
District	O
Mobile	O
Communication	O
Association	O
,	O
Sichuan	O
Changhong	O
Co.	O
,	O
Ltd.	O
,	O
Shanghai	O
SVA	O
(	O
Group	O
)	O
Central	O
Research	O
Institute	O
,	O
Zte	O
Communication	O
Co.	O
,	O
Ltd.	O
,	O
Zhongguancun	O
Hi-Tech	O
Industry	O
Association	O
,	O
established	O
the	O
AVS	B-Algorithm
industry	O
alliance	O
in	O
Beijing	O
.	O
</s>
<s>
The	O
organization	O
's	O
English	O
name	O
is	O
"	O
AVS	B-Algorithm
Industry	O
Alliance	O
"	O
(	O
AVSA	O
)	O
,	O
constituting	O
"	O
Three	O
Carriages	O
"	O
with	O
"	O
AVS	B-Algorithm
Workgroup	O
"	O
and	O
"	O
AVS	B-Algorithm
Patent	O
Pool	O
Management	O
Committee	O
"	O
.	O
</s>
<s>
The	O
first	O
generation	O
AVS	B-Algorithm
standard	O
included	O
Chinese	O
national	O
standard	O
“	O
Information	O
Technology	O
,	O
Advanced	O
Audio	O
Video	O
Coding	O
,	O
Part	B-Algorithm
2	I-Algorithm
:	O
Video	O
”	O
(	O
AVS1	B-Algorithm
for	O
short	O
,	O
GB	O
label：GB／T	O
20090.2-2006	O
)	O
and	O
“	O
Information	O
Technology	O
,	O
Advanced	O
Audio	O
Video	O
Coding	O
Part	O
16	O
:	O
Radio	O
Television	O
Video	O
”	O
(	O
AVS+	O
for	O
short	O
,	O
GB	O
label	O
:	O
GB/T	O
20090.16-2016	O
)	O
.	O
</s>
<s>
A	O
test	O
hosted	O
by	O
the	O
Radio	O
and	O
Television	O
Planning	O
Institute	O
of	O
the	O
State	O
Administration	O
of	O
Radio	O
,	O
Film	O
,	O
and	O
Television	O
(	O
SARFT	O
,	O
later	O
part	O
of	O
the	O
National	O
Radio	O
and	O
Television	O
Administration	O
)	O
shows	O
:	O
if	O
the	O
AVS1	B-Algorithm
bitrate	O
is	O
half	O
of	O
MPEG-2	B-Algorithm
standard	O
,	O
the	O
coding	O
quality	O
will	O
reach	O
excellent	O
for	O
both	O
standard	O
definition	O
or	O
high	O
definition	O
;	O
if	O
the	O
bitrate	O
is	O
less	O
than	O
1/3	O
,	O
it	O
also	O
reaches	O
good-excellent	O
levels	O
.	O
</s>
<s>
The	O
AVS1	B-Algorithm
standard	O
video	O
part	O
was	O
promulgated	O
as	O
the	O
Chinese	O
national	O
standard	O
in	O
February	O
2006	O
.	O
</s>
<s>
Around	O
this	O
time	O
,	O
AVS	B-Algorithm
was	O
considered	O
for	O
use	O
in	O
the	O
enhanced	O
versatile	O
disc	O
format	O
,	O
although	O
products	O
never	O
reached	O
the	O
market	O
.	O
</s>
<s>
During	O
the	O
May	O
7	O
–	O
11	O
,	O
2007	O
meeting	O
of	O
the	O
ITU-T	O
(	O
ITU	O
Telecommunication	O
Standardization	O
Sector	O
)	O
,	O
AVS1	B-Algorithm
was	O
one	O
of	O
the	O
standards	O
available	O
for	O
Internet	O
Protocol	O
television	O
(	O
IPTV	O
)	O
along	O
with	O
MPEG-2	B-Algorithm
,	O
H.264	B-Application
and	O
VC-1	B-Operating_System
.	O
</s>
<s>
On	O
June	O
4	O
,	O
2013	O
,	O
the	O
AVS1	B-Algorithm
video	O
part	O
was	O
issued	O
by	O
the	O
Institute	O
of	O
Electrical	O
and	O
Electronics	O
Engineers	O
(	O
IEEE	O
)	O
as	O
standard	O
IEEE1857-2013	O
,	O
</s>
<s>
AVS+	O
is	O
not	O
only	O
the	O
radio	O
,	O
film	O
and	O
television	O
industry	O
standard	O
GY/T	O
257.1-2012	O
“	O
Advanced	O
Audio	O
Video	O
Coding	O
for	O
Radio	O
and	O
Television	O
,	O
Part	B-Algorithm
1	I-Algorithm
:	O
Video	O
”	O
issued	O
by	O
the	O
SARFT	O
on	O
July	O
10	O
,	O
2012	O
,	O
but	O
also	O
the	O
enhanced	O
version	O
of	O
AVS1	B-Algorithm
.	O
</s>
<s>
The	O
second-generation	O
AVS	B-Algorithm
standard	O
included	O
the	O
series	O
of	O
Chinese	O
national	O
standard	O
“	O
Information	O
Technology	O
,	O
Efficient	O
Multi	O
Media	O
Coding	O
”	O
(	O
AVS2	B-Algorithm
)	O
.	O
</s>
<s>
AVS2	B-Algorithm
mainly	O
faces	O
the	O
transmission	O
of	O
extra	O
HD	O
TV	O
programs	O
,	O
</s>
<s>
The	O
SARFT	O
issued	O
AVS2	B-Algorithm
video	O
as	O
the	O
industry	O
standard	O
in	O
May	O
2016	O
,	O
and	O
as	O
the	O
Chinese	O
national	O
standard	O
on	O
December	O
30	O
,	O
2016	O
.	O
</s>
<s>
AVS2	B-Algorithm
was	O
published	O
by	O
the	O
Institute	O
of	O
Electrical	O
and	O
Electronics	O
Engineers	O
(	O
IEEE	O
)	O
as	O
standard	O
1857.4-2018	O
in	O
August	O
,	O
2019	O
.	O
</s>
<s>
A	O
test	O
showed	O
the	O
coding	O
efficiency	O
of	O
AVS2	B-Algorithm
more	O
than	O
doubled	O
that	O
of	O
AVS+	O
,	O
and	O
the	O
compression	O
rate	O
surpassed	O
the	O
international	O
standard	O
HEVC	O
(	O
H.265	O
)	O
.	O
</s>
<s>
Compared	O
with	O
the	O
first	O
generation	O
AVS	B-Algorithm
standard	O
,	O
the	O
second	O
can	O
save	O
half	O
transmission	O
bandwidth	O
.	O
</s>
<s>
AVS2	B-Algorithm
adopts	O
a	O
hybrid-coding	O
framework	O
,	O
and	O
the	O
whole	O
coding	O
process	O
includes	O
modules	O
such	O
as	O
intra-frame	O
prediction	O
,	O
inter-frame	O
prediction	O
,	O
transformation	O
,	O
quantization	O
,	O
inverse	O
quantization	O
and	O
inverse	O
transformation	O
,	O
loop	O
filter	O
and	O
entropy	O
coding	O
.	O
</s>
<s>
In	O
order	O
to	O
satisfy	O
the	O
requirements	O
of	O
HD	O
and	O
Ultra	B-Device
HD	I-Device
resolution	O
videos	O
for	O
the	O
compression	O
efficiency	O
,	O
AVS2	B-Algorithm
adopts	O
a	O
block	O
partition	O
structure	O
based	O
on	O
the	O
quadtree	B-Data_Structure
,	O
including	O
the	O
CU	O
(	O
Coding	O
Unit	O
)	O
,	O
PU	O
(	O
Prediction	O
Unit	O
)	O
and	O
TU	O
(	O
Transform	O
Unit	O
)	O
.	O
</s>
<s>
An	O
image	O
is	O
partitioned	O
into	O
LCU	O
(	O
Largest	O
CU	O
)	O
of	O
fixed	O
size	O
,	O
which	O
is	O
iterated	O
and	O
partitioned	O
into	O
a	O
series	O
of	O
CUs	O
in	O
the	O
form	O
of	O
quadtree	B-Data_Structure
.	O
</s>
<s>
Compared	O
with	O
the	O
traditional	O
macro	O
block	O
,	O
the	O
partition	O
structure	O
based	O
on	O
the	O
quadtree	B-Data_Structure
is	O
more	O
flexible	O
,	O
with	O
the	O
CU	O
size	O
extended	O
from	O
8×8	O
to	O
64×64	O
.	O
</s>
<s>
On	O
the	O
basis	O
of	O
AVS1	B-Algorithm
square	O
intra-frame	O
prediction	O
blocks	O
,	O
the	O
non-square	O
intra-frame	O
prediction	O
block	O
partition	O
is	O
added	O
.	O
</s>
<s>
Besides	O
CU	O
and	O
PU	O
,	O
AVS2	B-Algorithm
also	O
defines	O
a	O
transformation	O
unit	O
TU	O
for	O
the	O
prediction	O
of	O
residual	O
transformation	O
and	O
quantization	O
.	O
</s>
<s>
Compared	O
with	O
the	O
AVS1	B-Algorithm
and	O
H.264/AVC	B-Application
,	O
AVS2	B-Algorithm
designs	O
33	O
modes	O
for	O
the	O
intra-frame	O
prediction	O
coding	O
of	O
luminance	O
blocks	O
,	O
including	O
DC	O
prediction	O
mode	O
,	O
plane	O
prediction	O
mode	O
,	O
bilinear	O
prediction	O
mode	O
and	O
30	O
angel	O
prediction	O
modes	O
.	O
</s>
<s>
Compared	O
with	O
AVS1	B-Algorithm
,	O
AVS2	B-Algorithm
increases	O
the	O
maximum	O
quantity	O
of	O
candidate	O
reference	O
frames	O
to	O
4	O
,	O
so	O
as	O
to	O
adapt	O
to	O
the	O
multi-level	O
reference	O
frame	O
management	O
,	O
which	O
also	O
takes	O
full	O
advantage	O
of	O
the	O
redundant	O
space	O
of	O
the	O
buffer	O
.	O
</s>
<s>
In	O
order	O
to	O
satisfy	O
the	O
requirements	O
of	O
multiple	O
reference	O
frame	O
management	O
,	O
AVS2	B-Algorithm
adopts	O
a	O
kind	O
of	O
multi-level	O
reference	O
frame	O
management	O
mode	O
.	O
</s>
<s>
On	O
the	O
basis	O
of	O
AVS1	B-Algorithm
's	O
three	O
image	O
types	O
I	O
,	O
P	O
,	O
B	O
,	O
according	O
to	O
the	O
requirements	O
of	O
application	O
,	O
AVS2	B-Algorithm
adds	O
the	O
forward	O
multi-hypothesis	O
prediction	O
image	O
F	O
.	O
Aiming	O
at	O
the	O
video	O
surveillance	O
,	O
scene	O
play	O
and	O
other	O
specific	O
applications	O
,	O
AVS2	B-Algorithm
designs	O
scene	O
frames	O
(	O
Image	O
G	O
and	O
Image	O
GB	O
)	O
and	O
reference	O
scene	O
frame	O
S	O
.	O
</s>
<s>
In	O
order	O
to	O
fully	O
exert	O
the	O
performance	O
of	O
the	O
skip/direct	O
mode	O
of	O
Frame	O
B	O
,	O
AVS2	B-Algorithm
also	O
adopts	O
multi-direction	O
skip/direct	O
mode	O
under	O
the	O
premise	O
of	O
retaining	O
the	O
original	O
skip/direct	O
mode	O
of	O
Frame	O
B	O
:	O
two-way	O
skip/direct	O
mode	O
,	O
symmetrical	O
skip/direct	O
mode	O
,	O
backward	O
skip/direct	O
mode	O
and	O
forward	O
skip/direct	O
mode	O
.	O
</s>
<s>
AVS2	B-Algorithm
divides	O
the	O
multi-hypothesis	O
prediction	O
into	O
two	O
categories	O
,	O
namely	O
temporal	O
and	O
spatial	O
multi-hypothesis	O
mode	O
.	O
</s>
<s>
The	O
scene	O
frame	O
is	O
proposed	O
by	O
AVS2	B-Algorithm
based	O
on	O
the	O
surveillance	O
video	O
coding	O
method	O
of	O
background	O
modeling	O
.	O
</s>
<s>
When	O
the	O
surveillance	O
tool	O
is	O
opened	O
,	O
AVS2	B-Algorithm
will	O
apply	O
a	O
certain	O
frame	O
in	O
the	O
video	O
as	O
the	O
scene	O
image	O
frame	O
G	O
,	O
which	O
can	O
be	O
considered	O
as	O
a	O
long-term	O
reference	O
for	O
the	O
subsequent	O
images	O
.	O
</s>
<s>
AVS2	B-Algorithm
can	O
generate	O
the	O
scene	O
image	O
frame	O
GB	O
with	O
some	O
frames	O
in	O
the	O
video	O
,	O
and	O
frame	O
GB	O
can	O
also	O
be	O
applied	O
as	O
a	O
long-term	O
reference	O
.	O
</s>
<s>
In	O
order	O
to	O
simplify	O
the	O
motion	O
compensation	O
,	O
AVS2	B-Algorithm
adopts	O
an	O
8-tap	O
interpolation	O
filter	O
based	O
on	O
DCT	B-General_Concept
transformation	O
,	O
which	O
requires	O
only	O
one	O
filtering	O
,	O
and	O
supports	O
the	O
generation	O
of	O
higher	O
motion	O
vector	O
accuracy	O
than	O
1/4	O
pixel	O
.	O
</s>
<s>
Transformation	O
coding	O
in	O
AVS2	B-Algorithm
mainly	O
applies	O
integer	O
DCT	B-General_Concept
transformation	O
,	O
which	O
is	O
directly	O
performed	O
on	O
the	O
transformation	O
blocks	O
of	O
Size	O
4×4	O
,	O
8×8	O
,	O
16×16	O
,	O
32x32	O
.	O
</s>
<s>
For	O
one	O
transformation	O
block	O
with	O
dimension	O
greater	O
than	O
64	O
,	O
a	O
logical	O
transformation	O
LOT	O
is	O
adopted	O
to	O
conduct	O
the	O
wavelet	O
transformation	O
,	O
followed	O
by	O
the	O
integer	O
DCT	B-General_Concept
transformation	O
.	O
</s>
<s>
After	O
the	O
DCT	B-General_Concept
transformation	O
is	O
achieved	O
,	O
AVS2	B-Algorithm
will	O
conduct	O
the	O
second	O
4	O
x	O
4	O
transformation	O
for	O
the	O
4	O
x	O
4	O
blocks	O
with	O
low	O
frequency	O
coefficients	O
,	O
thus	O
further	O
to	O
reduce	O
the	O
correlation	O
between	O
coefficients	O
,	O
and	O
enable	O
the	O
energy	O
to	O
be	O
more	O
concentrated	O
.	O
</s>
<s>
The	O
AVS2	B-Algorithm
entropy	O
coding	O
divides	O
transformation	O
coefficients	O
into	O
CGs	O
(	O
Coefficient	O
Group	O
)	O
of	O
4	O
x	O
4	O
size	O
first	O
,	O
and	O
then	O
conducts	O
encoding	O
and	O
zigzag	O
scan	O
according	O
to	O
the	O
CGs	O
.	O
</s>
<s>
Binary	O
arithmetic	O
coding	O
and	O
two-dimensional	O
variable-length	O
coding	O
based	O
on	O
the	O
context	O
are	O
still	O
applied	O
in	O
the	O
AVS2	B-Algorithm
.	O
</s>
<s>
Loop	O
filter	O
modules	O
of	O
AVS2	B-Algorithm
contain	O
three	O
parts	O
:	O
deblocking	O
filter	O
,	O
adaptive	O
sample	O
point	O
offset	O
and	O
sample	O
compensation	O
filter	O
.	O
</s>
<s>
The	O
AVS2	B-Algorithm
adds	O
an	O
adaptive	O
filter	O
after	O
the	O
deblocking	O
filter	O
and	O
sample	O
offset	O
compensation	O
,	O
a	O
Wiener	O
filter	O
with	O
7×7	O
cross	O
plus	O
3×3	O
square	O
centrosymmetry	O
,	O
which	O
applies	O
the	O
original	O
undistorted	O
image	O
and	O
encoding	O
reconstructed	O
image	O
to	O
figure	O
out	O
the	O
least	O
square	O
filter	O
coefficient	O
,	O
and	O
conduct	O
filtering	O
on	O
the	O
decoding	O
reconstructed	O
image	O
,	O
thus	O
to	O
reduce	O
the	O
compression	O
distortion	O
in	O
the	O
decoding	O
image	O
,	O
and	O
enhance	O
the	O
quality	O
of	O
the	O
reference	O
image	O
.	O
</s>
<s>
FFmpeg	B-Library
V6	I-Library
can	O
make	O
use	O
of	O
the	O
uAVS3d	O
library	O
for	O
AVS3-P2/IEEE1857.10	O
video	B-Application
decoding	I-Application
.	O
</s>
<s>
Competing	O
codec	B-General_Concept
from	O
MPEG	O
:	O
</s>
<s>
MPEG-H	B-Algorithm
-	I-Algorithm
VVC	I-Algorithm
-	I-Algorithm
H.266	I-Algorithm
are	O
still	O
not	O
officially	O
available	O
Out	B-General_Concept
of	I-General_Concept
the	I-General_Concept
Box	I-General_Concept
within	O
FFmpeg	B-Library
.	O
</s>
<s>
uAVS3d	O
is	O
an	O
Open	B-License
Source	I-License
and	O
cross-platform	B-Operating_System
AVS3	O
decoder	O
.	O
</s>
<s>
supports	O
to	O
compile	B-Language
for	O
Android/iOS/Windows/Linux/MacOS	O
systems	O
.	O
</s>
<s>
optimized	O
for	O
ARMv7/ARMv8/SSE4/AVX2	O
chips	O
.	O
</s>
<s>
The	O
uAVS3	O
video	B-Application
codec	I-Application
has	O
supported	O
x86	B-Operating_System
and	O
ARM	B-Architecture
platforms	O
,	O
and	O
has	O
been	O
tested	O
and	O
verified	O
on	O
the	O
Kunpeng	O
processor	O
.	O
</s>
<s>
uAVS3e	O
is	O
an	O
Open	B-License
Source	I-License
AVS3	O
encoder	O
.	O
</s>
<s>
supports	O
to	O
compile	B-Language
for	O
Windows/Linux	O
systems	O
.	O
</s>
<s>
optimized	O
for	O
SSE4/AVX2	O
chips	O
.	O
</s>
<s>
10bit	O
encoding	O
on	O
all	O
supported	O
platforms	O
(	O
without	O
SIMD	B-Device
)	O
.	O
</s>
<s>
The	O
uAVS3	O
codec	B-General_Concept
has	O
supported	O
x86	B-Operating_System
and	O
ARM	B-Architecture
platforms	O
,	O
and	O
has	O
been	O
tested	O
and	O
verified	O
on	O
the	O
Kunpeng	O
processor	O
.	O
</s>
<s>
It	O
is	O
a	O
free	B-Application
software	I-Application
under	O
the	O
terms	O
of	O
the	O
BSD	O
3-clause	O
license	O
.	O
</s>
<s>
It	O
is	O
a	O
free	B-Application
software	I-Application
under	O
the	O
terms	O
of	O
the	O
BSD	O
4-clause	O
license	O
.	O
</s>
<s>
Subsequently	O
,	O
AVS2	B-Algorithm
Ultra	B-Device
HD	I-Device
real-time	O
video	O
encoder	O
and	O
mobile	O
HD	O
encoder	O
were	O
announced	O
.	O
</s>
<s>
OpenAVS2	O
is	O
a	O
set	O
of	O
audio	O
and	O
video	O
coding	O
,	O
transcoding	O
and	O
decoding	O
software	O
based	O
on	O
the	O
AVS2	B-Algorithm
standard	O
.	O
</s>
<s>
and	O
are	O
open-source	B-License
encoder	O
and	O
decoder	O
published	O
by	O
Peking	O
University	O
Video	O
Coding	O
Laboratory	O
(	O
)	O
based	O
on	O
AVS2-P2/	O
video	O
coding	O
standard	O
,	O
which	O
is	O
offered	O
under	O
either	O
version	O
2	O
of	O
the	O
GNU	B-License
General	I-License
Public	I-License
License	I-License
(	O
GPL	B-License
)	O
or	O
a	O
commercial	O
license	O
.	O
</s>
<s>
and	O
it	O
can	O
make	O
use	O
of	O
the	O
xAVS2	O
library	O
for	O
AVS2-P2/IEEE1857.4	O
video	O
encoding	O
.	O
</s>
<s>
Libdavs2	O
and	O
Libxavs2	O
are	O
under	O
the	O
GNU	B-License
Public	I-License
License	I-License
Version	O
2	O
or	O
later	O
.	O
</s>
