theora - Jul 2009 - ffmpeg2theora: issue converting from a png image list

Hi Jan, all

I am trying to generate a video introduction with a fade-in / fade-out
effect. To do this with ffmpeg2theora only, I generated a list of PNG
images: img001.png, img002.png... img299.png

The ffmpeg2theora man page says:

Encode a series of images:
         ffmpeg2theora frame%06d.png -o output.ogv

That's exactly what I tried with the latest 0.24 release of ffmpeg2theora:

ffmpeg2theora-0.24.linux32.bin img-%03d.png -o output.ogv
Input #0, image2, from 'img-%03d.png':
  Duration: 00:00:12.00, start: 0.000000, bitrate: N/A
    Stream #0.0: Video: png, yuv420p, 1280x720, 25 tbr, 25 tbn, 25 tbc
  Resize: 1280x720
Segmentation fault

The same problem happens with the release in Ubuntu 9.04, and with the
0.22 release. I haven't tried earlier releases yet.

To reproduce this issue, you can get the images from the tarball on
http://free-electrons.com/issues/ffmpeg2theora-jul15-2009/

I would like to avoid using mencoder + ffmpeg2theora to do the job
(unless I manage to generate the theora video directly from mencoder).

I hope this helps to fix this issue... Don't hesitate to get back to me
if you have any question.

Cheers,

Michael.

-- 
Michael Opdenacker, Free Electrons
Kernel, drivers and embedded Linux development,
consulting, training and support.
http://free-electrons.com
+ 33 621 604 642

On Wed, Jul 15, 2009 at 3:24 PM, Michael
Opdenacker<michael at free-electrons.com> wrote:
> Hi Jan, all
>
> I am trying to generate a video introduction with a fade-in / fade-out
> effect. To do this with ffmpeg2theora only, I generated a list of PNG
> images: img001.png, img002.png... img299.png
>
> The ffmpeg2theora man page says:
>
> Encode a series of images:
> ? ? ? ? ffmpeg2theora frame%06d.png -o output.ogv
>
> That's exactly what I tried with the latest 0.24 release of
ffmpeg2theora:
>
> ffmpeg2theora-0.24.linux32.bin img-%03d.png -o output.ogv
> Input #0, image2, from 'img-%03d.png':
> ?Duration: 00:00:12.00, start: 0.000000, bitrate: N/A
> ? ?Stream #0.0: Video: png, yuv420p, 1280x720, 25 tbr, 25 tbn, 25 tbc
> ?Resize: 1280x720
> Segmentation fault
Indeed, the same happens for me too. But try this:

ffmpeg -i img-%03d.png -r 15 -vcodec copy output.avi
ffmpeg2theora output.avi -o output.ogv

this works for me. Could someone please clarify if I am loosing any
quality in the first step? E.g. notice the "-vcodec copy" thing, also
it's *really* fast, e.g. just like if it just copied the images
without any compression. But I might be wrong.


Ondrej

> ffmpeg2theora-0.24.linux32.bin img-%03d.png -o output.ogv
> Input #0, image2, from 'img-%03d.png':
>   Duration: 00:00:12.00, start: 0.000000, bitrate: N/A
>     Stream #0.0: Video: png, yuv420p, 1280x720, 25 tbr, 25 tbn, 25 tbc
>   Resize: 1280x720
> Segmentation fault
Can you try again with the latest svn ?
It works for me, and I'd fixed a crash a month or two ago that might
well be this one.