#11462: Cannot embed .scc file into .mp4 using -c:s copy
-------------------------------------+-------------------------------------
Reporter: Zach | Owner: (none)
Type: enhancement | Status: new
Priority: important | Component:
| undetermined
Version: git-master | Resolution:
Keywords: | Blocked By:
Blocking: | Reproduced by developer: 0
Analyzed by developer: 0 |
-------------------------------------+-------------------------------------
Comment (by softworkz):
Your diagrams are accurate with one exception: The splitcc filter
currently employs the CEA-608 captions decoder, so it outputs text
subtitles (ass). In the same way would a mergecc filter employ a CEA-608
(optionally also 708) encoder.
My idea for scenarios where the original CC bitstream is desired to be
used, the splitcc filter could simply attach the original CC data to the
subtitle frames (AVFrame, just like for video and audio - that's the big
change that subtitle filtering makes).
**splitcc**
That means: splitcc takes video frames as input which have the CC data
attached as side-data. It decodes the CC and outputs (1) the unchanged
video frames and (2) text subtitles frames (ass). Now we could simply make
a copy of the CC side data from the video frames and attach it to the text
subtitle frames.
This opens a wide range of possibilities:
- The subtitles output could be further processed as text in subsequent
filters
e.g. modified, styles changes, overlaid/burn-in on the video, onverted
to graphic subtitles, etc.
- The subtitles output can leave filtering and converted to a specific
text-sub format and muxed into a stream or written to disk separately
- The subtitles ouput frames (with CC data) could be attached to the same
or a different video stream's frames (again by copying the side data) ==>
embedcc (or mergecc)
- The subtitles ouput frames (with CC data) could be written to disk as
scc or other data
**mergecc**
The regular operation for mergecc would be to take text subtitle frames
are input and employ a CEA608 (or also 708) encoder which encodes the text
subs as CC and attaches it as side data to the video frames.
Probably it makes sense to have a separate filter for working on the CC
data of text subtitle frames, so maybe like
**embedcc**
Opposed to mergecc it would take the attached side data from the text
subtitle frames and well - do something to make it a valid DTVCC stream
when attaching CC side data to the video frames.
I'm not sure whether this is easily possible...
I think it's more safe to encode from ASS text subs, but if Devin says he
knows how to do it, then it could work that way.
--
Ticket URL: <https://trac.ffmpeg.org/ticket/11462#comment:33>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker
_______________________________________________
FFmpeg-trac mailing list
FFmpeg-trac@avcodec.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-trac
To unsubscribe, visit link above, or email
ffmpeg-trac-requ...@ffmpeg.org with subject "unsubscribe".
