On 26/06/2025 16.51, Carl Zwanzig wrote:
On 6/26/2025 10:32 AM, Mark Filipak wrote:
Z. please stop implying that I don't search for answers.
At least an appearance of doing that, the first email of this thread did not; maybe tell us what you
found, where, and why it doesn't answer the questions. Even question your own assumptions, like "it
doesn't look like a bit map but everyone says it is" - is everyone else wrong or are you missing
some essential info or looking at it in the wrong way? (That's basic scientific method.)
On 6/26/2025 12:39 PM, Mark Filipak wrote:
but I'm going by what I see, not by the crap on the Internet.
'crap' is a strange description for something the Library of Congress published, did you read it?
Maybe you should send them a correction.
Do you have a URL? "Library of Congress"? About subtitles? I want to read it.
_Another_ quick search turns up a quite in-depth description of the subtitle stream*, including the
image itself. Among other things, the lines are interlaced -and- the pixels are run-length encoded,...
Note "run-length encoded". Isn't that what I said?
...plus color info is included (from a palette defined elsewhere) -
That palette is really what I'm after. *.idx always shows a 16-element, triple-byte (rgb? yuv?)
palette, but rewriting that has no effect on the subtitles that are OCRed. *.idx also can have an
optional 4-element, triple-byte palette with a 4-element transparency nibble, but it appears to also
be ignored by subtitle tools. If I could find that palette in the stream, maybe I could
change-correct it.
_______________________________________________
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
