headers
Víctor Paesa | 1 Jan 10:56

Re: composing multiple video layers into one

Hi,

Rich Rodecker said:
> Hello,
> I am trying to figure out a way to compose multiple video 'layers' into
> one output file via command line (or, at the very least, overlay a png
> over the video, where the png file is always going to be much larger
> than the video frame size).  As far as I can tell, this cannot be done
> with ffmpeg alone, correct? I have been doing a ton of research and have
> looked at many, many options such as mencoder, kino, cinelerra, etc, but
> everything's kind of becoming noise at this point.  If anyone can help
> me out with details, that would be awesome, but just pointing me in the
> right direction would be great too.
>
> Right now my process is:
>   1.  Convert video to jpg sequence
>   2.  Create new, composited frame sequence images in GD.
>   3.  Create a new video file from the composited images using ffmpeg.
>
> It pretty much works, but is time (and resource) consuming...i figure
> there has got to be a better way.  I think the main monkey wrench here
> is that the size of the image/video I want to compose is different than
> the original video.

Maybe the PiP vhook in
http://wiki.multimedia.cx/index.php?title=Interesting_Patches

Not sure to understand your scenario, though. Could you elaborate
on your step 2?
(Continue reading)

Permalink | Reply | 5 comments |
headers
jersingsu.tw | 1 Jan 12:51
Picon
Favicon

Re: using libavcodec on cygwin, and ability to use ffmpeg to edit video

Hi:
Sorry, I didn't find any information about how to
compile WinCE code using arm-wince-mingw32ce and
Cygwin.

Could you tell me other useful information about how
to compile WinCE code ?

Thanks !!!
--Kirk

--- V獳tor Paesa <wzrlpy <at> arsystel.com> 說：

> Hi,
> 
> jersingsu.tw said:
> > Dear~
> >
> > I used on MinGW on Windows:
> > ./configure --cross-compile
> >
>
--cross-prefix=/usr/local/wince/mingw32ce/arm-wince-mingw32ce/bin/
> > --arch=armv4l --enable-memalign-hack
> > --prefix=/usr/local/wince --enable-shared
> > --disable-static
> >
> > config.err told me:
> > BEGIN /tmp/ffmpeg-conf-20745-1216-24186.c
> > END /tmp/ffmpeg-conf-20745-1216-24186.c
(Continue reading)

Permalink | Reply | 0 comments |
headers
jersingsu.tw | 2 Jan 09:00
Picon
Favicon

Re: using libavcodec on cygwin, and ability to use ffmpeg to edit video

I build ffmpeg :
$ ./configure --cross-compile
--cross-prefix=/usr/local/wince/mingw32ce/arm-win
ce-mingw32ce/bin/ --arch=armv4l --target-os=mingw32
--enable-memalign-hack --en
able-shared --disable-static
--extra-cflags=-mno-cygwin --extra-libs=-mno-cygwi
n --disable-ipv6 --prefix=/usr/local/wince --log=yes
--disable-encoders --disab
le-decoders --disable-parsers --disable-muxers
--disable-demuxers --enable-pars
er=h263 --enable-encoder=h263 --enable-decoder=h263
--enable-muxer=h263 --enabl
e-demuxer=h263 --disable-ffserver --disable-ffplay
--disable-ffmpeg --disable-n
etwork

config.err:
.....................
BEGIN /tmp/ffmpeg-conf-8681-3968-4213.c
END /tmp/ffmpeg-conf-8681-3968-4213.c
/usr/local/wince/mingw32ce/arm-wince-mingw32ce/bin/gcc
-mno-cygwin -fomit-frame-pointer -c -o
/tmp/ffmpeg-conf-29704-3968-10397.o
/tmp/ffmpeg-conf-8681-3968-4213.c
gcc: error trying to exec 'cc1': execvp: No such file
or directory
C compiler test failed.

It's the same error after I copy the cc1 to the same
(Continue reading)

Permalink | Reply | 16 comments |
headers
Michel Bardiaux | 2 Jan 11:25
Picon

Re: using libavcodec on cygwin, and ability to use ffmpeg to edit video

jersingsu.tw a écrit :
> I build ffmpeg :

[snip]

> 
> Could you give me someone hint to solve this issue ?

*NO*. Read what Victor wrote.

> Thanks a lot !!!
> --Kirk
> --- V?tor Paesa <wzrlpy <at> arsystel.com> ?:
> 
[snip]

>> Please read and follow
>> http://ffmpeg.mplayerhq.hu/mailinglists.html
>> (hint: read specially top-posting/hijacking in the
>> first phrase there)
>>
>> Regards,
>> V?tor
>>

--

-- 
Michel Bardiaux
R&D Director
T +32 [0] 2 790 29 41
F +32 [0] 2 790 29 02
(Continue reading)

Permalink | Reply | 15 comments |
headers
Hasseeb Ashfaq | 2 Jan 12:00
Picon
Favicon

Question Regarding Optimizing FFMpeg on Blackfin

Hello,
  First I have some problem regarding login in.
  I registered and then confirmed the confirmation email and I recieved the welcome message as well. But then
i wanted to login to list i kept recieving error message which is am pasting below.
  Error  ffmpeg-user roster authentication failed. 

---------------------------------

  ffmpeg-user list run by michaelni at users.sourceforge.net, baptiste.coudurier at free.fr
ffmpeg-user administrative interface (requires authorization)
Overview of all mplayerhq.hu mailing lists        so could you plz help me in this regard.        Second I have a question
regarding FFMpeg implementation on Blackfin. I am mailing it to you if you could post it on the list.        I am
trying to implement FFMPEG 2 Decoder on BF561.
My BF561 is running at 600MHZ, my decoder is outputing 100 frames in 13 
seconds. If i enable caching it takes 3.8 seconds (just less than real time as 
real time requires 4secs to decode 100 frames) but if I enable PPI with it then 
it takes 6 seconds.
So any idea how could i make it run faster then this?
worth mentioning that if i use mannual caching(then DMA runs at 133MHZ which is 
slower than processor and net result, DMA slows the processor when enabled)
If i enable CPLB option in caching, and select it for write back, I recieve 
exception.... why? I dont know
I have already turned on optimization and set it to 100%. 
I have used Profiling, and profiling shows bulk of time is used by functions
1. ff_bfin_idct (it is written in assembly and all its parameters are in L1 
memory so i believe it cant go any faster than this.
2. ff_bfin_put_pixels8uc
3. ff_bfin_put_pixels16uc
4. ff_bfin_z_put_pixels16_xy2
5. avg_pixels8_xy2_c
(Continue reading)

Permalink | Reply | 0 comments |
headers
Michel Bardiaux | 2 Jan 17:25
Picon

Re: avcodec_decode_audio2 doesn't decode small buffers

Thomas Rehaag a écrit :
> Hi Michel,
> 
> I took your line 
>> ffmpeg -dump -v 2 -i tvshort.mpg -y jul.mpg | & tee -a notes
> and modified it for my in.mpeg:
> 
> ffmpeg -dump -v 2 -i in.mpg -y jul.mpg
> 
> And this delivers a clear audio buffer size of 192 bytes.
> Then I've tried my code, and avcodec_decode_audio2 decodes 192 bytes at
> once, but av_read_packet e.g. reads 2037 bytes the first time it's
> called.
> And the ffmpeg dump shows the decoding of about 11 audio frames of 192
> bytes, before the next video frame appears. About the same size, isn't
> it? 

So, to summarize: the problem is neither in the input file, nor in 
ffmpeg&libs, it must be in your code. Can you send me a zip of your code 
plus a Makefile, so I can try it? (As much as I can, my C++ is very --)

> 
> Can I have done something wrong, compiling ffmpeg, so that
> av_read_packet does something wrong? 

Doesnt seem likely, but you never know, do a distclean.

> (Hope I've got you right, that it
> should deliver handy little packages of 192 bytes in case of in.mpg)
(Continue reading)

Permalink | Reply | 6 comments |
headers
Vinod G | 2 Jan 19:59
Picon

persistent http connection

Is it possible to configure the ffmpeg's http client to use persistent
connections?

Thanks,
Vinod.
Permalink | Reply | 0 comments |
headers
Thomas Rehaag | 2 Jan 21:22
Picon
Favicon

Re: avcodec_decode_audio2 doesn't decode small buffers

Hi Michel,

> So, to summarize: the problem is neither in the input file, nor in
> ffmpeg&libs, it must be in your code. Can you send me a zip of your code
> plus a Makefile, so I can try it? (As much as I can, my C++ is very --)

The code is here:
http://www.ismism.de/xc/twoinone.zip
don't remember, if I mentioned before - the program stitches together
two input videos. E.g.: two times 640*480 -> 1280*480.
I don't think, you'll have to know much about c++ to read the code. Else
I'll have to write an example code using C only.
And also I hope, Visual C++ exported a correct makefile.

And - the audio problem is my minor problem:

- The program can't decode wmv files at all. I set the output's
framerate hardcoded to 30 (not in the uploaded code), cause the output
codec will not be created at 1000 frames/s. Then the first packet of
3877 bytes seems to be decoded well. The second time, av_read_packet
delivers 32397 bytes and avcodec_decode_video returns -1.

- Decoding and encoding MPEG1 files seems to work quite well at the
first sight. My TV testfiles, including sound, 

work fine in between. 
But the files, my main program encoded with the help of mencoder:
mencoder.exe a.wmv -o a.mpg -of mpeg -mpegopts
format=mpeg1:tsaf:muxrate=2000 -oac lavc -ovc lavc -lavcopts
acodec=mp2:abitrate=224:vcodec=mpeg1video:vbitrate=2304:keyint=15:mbd=2
(Continue reading)

Permalink | Reply | 5 comments |
headers
Kyle Mallory | 2 Jan 22:33
Picon
Favicon
Gravatar

trouble encoding with libavcodec

I am in something of a bind.  I am writing a FFMPEG plugin for a
professional, high-end NLE application, running on Linux.  The decoder
seems to be pretty robust, with the ability to compare the visual clips
in the NLE with the output of ffplay, etc.  However, on the encoder
side, where the frames are originating with the application, I am unable
to do comparisons between ffmpeg encode options, and libavcodec options.

The problem I'm having is that the files that I am encoding are less
than reliable.  About 60% of the time, something I encode will play back
with recognizable video, but with errors such as color noise, green
artifacts, etc.  With some codecs, I get an immediate SIGSEGV, or other
fault.

I'm trying to find the best way to debug these issues, but I'm not
having much luck.  Ultimately, I'm trying to find a way to continue to
work on this project, and get it running, realizing I'm a lone tree in a
desert, isolated by this commercial application.

Also worth noting, is that I'm running an old FFMPEG/libav*.  Since its
a commercial application, I am resolved to using the latest distribution
build as my base requirements.  Using the latest version made things run
a little more smoothly, but still not perfect.

I've written a handful of routines as a basic API around libav*, to help
in my development and testing, so I can validate results without the
need for the commercial application, but it has only gotten me so far.

I'd like to post the API and test routines, and see if anyone else has
problems, or can review the code for issues.  It's about 1200 lines,
spread over about 5 files (including headers).  The API is fairly light,
(Continue reading)

Permalink | Reply | 0 comments |
headers
Thomas Rehaag | 2 Jan 23:29
Picon
Favicon

Re: avcodec_decode_audio2 doesn't decode small buffers


Thomas Rehaag schrieb:
> 
> Hi Michel,
> 
> > So, to summarize: the problem is neither in the input file, nor in
> > ffmpeg&libs, it must be in your code. Can you send me a zip of your code
> > plus a Makefile, so I can try it? (As much as I can, my C++ is very --)
> 
> The code is here:
> http://www.ismism.de/xc/twoinone.zip
> don't remember, if I mentioned before - the program stitches together
> two input videos. E.g.: two times 640*480 -> 1280*480.
> I don't think, you'll have to know much about c++ to read the code. Else
> I'll have to write an example code using C only.
> And also I hope, Visual C++ exported a correct makefile.
> 
> And - the audio problem is my minor problem:
> 
> - The program can't decode wmv files at all. I set the output's
> framerate hardcoded to 30 (not in the uploaded code), cause the output
> codec will not be created at 1000 frames/s. Then the first packet of
> 3877 bytes seems to be decoded well. The second time, av_read_packet
> delivers 32397 bytes and avcodec_decode_video returns -1.
> 
> - Decoding and encoding MPEG1 files seems to work quite well at the
> first sight. My TV testfiles, including sound,
> 
> work fine in between.
> But the files, my main program encoded with the help of mencoder:
(Continue reading)

Permalink | Reply | 1 comment |

Gmane