<br>thank you<br><br>in fact, my purpose is simple<br>I just want to encode a sequence into four or more layers<br>and obtain sizes of each frame(or a layer segment)<br>time duration of each frame<br>and use such information for network simulation<br>
and then compute PSNR <br>is there any good method?<br><br><br>you said :<br>Using the --b-pyramid option, it is also probably possible to generate<br>
a 3 layers stream, the first layer containning I/P frames, the second<br>
reference B frames and the third non-reference B frames.<br>_______________________________________________<br><br>can we treat I frame as the first layer<br>P frame as the second one<br>B frame as the third while the third non-refrence B(let's call it b) as the forth layer?<br>
<br>besides, I didn't see any difference between single layer and such temporal SVC<br>we just regard different frames as different layers <br>I guess we can still treat these frames as in the same single layer?<br><br>
<br><br><br><div class="gmail_quote">On Fri, Jun 19, 2009 at 9:24 AM, Loic Le Loarer <span dir="ltr"><<a href="mailto:lll%2Bvlc@m4x.org">lll+vlc@m4x.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi,<br>
<div class="im"><br>
On Thursday June 18 2009 at 12:30:26 PM +0200, 武南龚 wrote:<br>
> thank you so much. I'm a little puzzled by the sentence below, can you give<br>
> me some explanation.<br>
><br>
> Yes, temporal scalability is just a single layer with non-reference<br>
> frames at regular intervals.<br>
> _______________________________________<br>
><br>
> besides, x264 doesn't support SVC, can I treat the single layer coded video<br>
> as a temporal scalable video coded one?<br>
<br>
</div>Yes, you can do that if the coded stream has the correct properties. For<br>
example, do be a correct temporal scalable video sequence with 2 layers,<br>
the base layer at 15fps and the second at 30fps, it is enough to have<br>
the following sequence of pictures in display order (using old MPEG2<br>
style for references):<br>
<br>
I B P B P B P...<br>
<br>
The base layer contains only the I and P pictures, the second layer adds<br>
the B pictures.<br>
<br>
Using x264, it is possible to generate such a sequence using the<br>
following options (or similar):<br>
<br>
--bframes 1 --b-adapt 0 --ref 2<br>
<br>
Using the --b-pyramid option, it is also probably possible to generate<br>
a 3 layers stream, the first layer containning I/P frames, the second<br>
reference B frames and the third non-reference B frames.<br>
<br>
Once you have the generated the sequence using x264, you can turn it to<br>
a real SVC sequence with temporal scalability by just adding the correct<br>
SEI messages and may be changing some SPS or PPS parameters. I don't<br>
know any tools whch can to this, but it isn't difficult to write this,<br>
it could even be done directly in x264 and proposed for inclusion I<br>
guess if there is interess in such a feature.<br>
<br>
Best regards.<br>
<font color="#888888"><br>
--<br>
Loic<br>
<br>
Only a very small fraction of our DNA does anything; the rest is all<br>
comments and ifdefs.<br>
</font><div><div></div><div class="h5"><br>
> and is there any other tools similar to x264 that support SVC?<br>
><br>
> thank you!<br>
><br>
><br>
><br>
> 2009/6/18 Måns Rullgård <<a href="mailto:mans@mansr.com">mans@mansr.com</a>><br>
><br>
> > 武南龚 <<a href="mailto:nangergong@gmail.com">nangergong@gmail.com</a>> writes:<br>
> ><br>
> > > hi,all<br>
> > ><br>
> > > I'm using x.264 in linux now, I want to know which command line<br>
> > argument<br>
> > > decides whether the encoding is single layer video coding or scalable<br>
> > video<br>
> > > coding?<br>
> > > and how to encode videos using different scalable video coding (temporal<br>
> > > ,spatial and PSNR)?<br>
> ><br>
> > SVC is not supported in x264.<br>
> ><br>
> > > besides, is there any difference between single layer video coding and<br>
> > > temporal scalable video coding? I thought they are the same<br>
> ><br>
> > Yes, temporal scalability is just a single layer with non-reference<br>
> > frames at regular intervals.<br>
> ><br>
> > --<br>
> > Måns Rullgård<br>
> > <a href="mailto:mans@mansr.com">mans@mansr.com</a><br>
> > _______________________________________________<br>
> > x264-devel mailing list<br>
> > <a href="mailto:x264-devel@videolan.org">x264-devel@videolan.org</a><br>
> > <a href="http://mailman.videolan.org/listinfo/x264-devel" target="_blank">http://mailman.videolan.org/listinfo/x264-devel</a><br>
> ><br>
><br>
><br>
><br>
> --<br>
> 此致<br>
> 敬礼<br>
><br>
> 龚武南<br>
><br>
> GongWunan<br>
><br>
> School of Electronic and Information Engineering<br>
> Xi'an Jiaotong University,<br>
> Xi'an 710049, People's Republic of China<br>
> ^<br>
> Computer and Communication Networks Engineering<br>
> Politecnico di Torino,<br>
> TORINO 10138,Italy<br>
> Tel:+39-3881278481<br>
<br>
> _______________________________________________<br>
> x264-devel mailing list<br>
> <a href="mailto:x264-devel@videolan.org">x264-devel@videolan.org</a><br>
> <a href="http://mailman.videolan.org/listinfo/x264-devel" target="_blank">http://mailman.videolan.org/listinfo/x264-devel</a><br>
<br>
_______________________________________________<br>
x264-devel mailing list<br>
<a href="mailto:x264-devel@videolan.org">x264-devel@videolan.org</a><br>
<a href="http://mailman.videolan.org/listinfo/x264-devel" target="_blank">http://mailman.videolan.org/listinfo/x264-devel</a><br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br> 此致<br> 敬礼<br> <br> 龚武南<br><br>GongWunan<br>
<br>School of Electronic and Information Engineering <br>Xi'an Jiaotong University, <br>Xi'an 710049, People's Republic of China<br>^<br>Computer and Communication Networks Engineering <br>Politecnico di Torino,<br>
TORINO 10138,Italy<br>Tel:+39-3881278481<br>