Re: [Opensvcdecoder-support] SVC standalone display timing

[Médéric B. <Med...@in...>]

Dear Alberto,

In case of different frame rate between two layers, pictures of the 
lower temporal layer are displayed several times to keep the same frame 
rate between all layers.
In a nutshell, if there is two layers (base layer at 12.5fps and one 
enhancement at 25fps), the frame of the base layer are displayed twice 
to keep the same number of displayed frame between all layers.

When decoder_svc_VideoParameter_ImgToDisplay is equals to 2, there is a 
temporal scalability in another layer, so the frame displayed is a copy.

decoder_svc_VideoParameter_ImgToDisplay can also be equals to 0 (no 
frame available) and to 1 (frame to display).

Regards,
Médéric



Le 19/09/2011 20:21, alberto alvarez gonzález a écrit :
> Thank you Mickael.
>
> I supected the timing was left to muxer, thank you for making it 
> straight to me.
> I can manage with -fps as input for now, without bringing a container 
> into play yet.
>
> What about the " decoder_svc_VideoParameter_ImgToDisplay" parameter?
> Why can it take two values and in the case of double frames, which 
> ones should I consider?
>
> Thanks for the quick response.
> Best regards
> Alberto Alvarez
>
>
> ------------------------------------------------------------------------
> CC: ope...@li...
> From: Mic...@in...
> Subject: Re: [Opensvcdecoder-support] SVC standalone display timing
> Date: Mon, 19 Sep 2011 19:49:05 +0200
> To: alb...@ho...
>
> A raw stream does not contain timing information. For any timing 
> information, you need a container mp4, mp2ts, Mkv. That's why you 
> going as fast as possible.
>
> Mickael
>
> Sent by my iPhone
>
> Le 19 sept. 2011 à 19:29, alberto alvarez 
> gonzález<alb...@ho... 
> <mailto:alb...@ho...>> a écrit :
>
>     Hello
>
>     I'm opening a new thread with more questions. Your help has
>     been invaluable so far.
>
>     Regarding the display timing, yesterday's svn showed the video as
>     fast as it was processed.
>     I have come across a quick workaround which pretty much works for
>     me. It is based on SDL_Delay function which forces the display
>     to wait 1/FPS. Given the dynamic fps nature of SVC I have thought
>     using NAL indexes MaxTemporalId and TempToDisplay to compute
>     the fraction of original fps (the max) that is used (linked to
>     -tempId input). Roughly:
>              fps_start_time=SDL_GetTicks();
>
>               SDL_Display(16, XDIM, YDIM, Y, U, V);
>
>             
>     current_fps=FPS>>(DqIdNextNal_Nal_o->MaxTemporalId-DqIdNextNal_Nal_o->TempToDisplay);
>              SDL_Delay((unsigned
>     int)1000.0/current_fps-(SDL_GetTicks()-fps_start_time));
>
>     However, I experienced two issues.
>     1- Can't find FPS of the sequence in any header of raw sequences
>     ¿is it there yet? For now is an input parameter.
>     2- With a fps reduction factor of 2 I should expect half the
>     frames. However, it shows ¿repeated? all the frames
>     Depending on decoder_svc_VideoParameter_ImgToDisplay value ([1,2])
>     there are two paths possible and in my example, the
>     frames are displayed from those code sequences alternatively. I
>     think the frames are repeated.
>
>     What are the implications of that parameter? Am I getting it right?
>
>     Any help would be appreciated.
>
>     Best Regards,
>
>     Alberto Alvarez
>
>     ------------------------------------------------------------------------------
>     BlackBerry&reg; DevCon Americas, Oct. 18-20, San Francisco, CA
>     Learn about the latest advances in developing for the
>     BlackBerry&reg; mobile platform with sessions, labs & more.
>     See new tools and technologies. Register for BlackBerry&reg;
>     DevCon today!
>     http://p.sf.net/sfu/rim-devcon-copy1
>
>     _______________________________________________
>     Opensvcdecoder-support mailing list
>     Ope...@li...
>     <mailto:Ope...@li...>
>     https://lists.sourceforge.net/lists/listinfo/opensvcdecoder-support
>
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2dcopy1
>
>
> _______________________________________________
> Opensvcdecoder-support mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/opensvcdecoder-support


-- 

Médéric Blestel
Ingénieur de Recherche / Research Engineer
Institut d' Electronique et de Télécommunications de Rennes (IETR)
UMR CNRS 6164

Tél : +33 (2) 23 23 85 67
Fax : +33 (2) 23 23 82 62


IETR/Groupe Image
INSA DE RENNES
20 AVENUE DES BUTTES DE COESMES
CS 70 839
35 708 RENNES CEDEX 7