[Ffmpeg-devel] tiff encoder (qualification task for GSoC)

[Justin Ruggles]

Kostya wrote:
>>It would be great to have JPEG compression here too.  There is a
>>technical note from Adobe on the correct way to do JPEG-in-TIFF.  LJPEG
>>is especially common, specifically for digital still cameras.  It might
>>be a bit complex trying to interface with the mjpeg encoder though.
> 
> 
> Hmm, I think Lavc should be able to decode its own output and JPEG-in-TIFF
> is not unlike DV-in-(AVI|MOV).
> 
> IIRC, there are two variants: JPEG components stored in TIFF tags separately
> and whole JPEG-in-JFIF-in-TIFF variant. First seems to be logical at least.
> Which one is used?

The correct variant, which is supported by libtiff, is to compress each
strip independently.  They strips can share tables through the
JpegTables tag or they can use their own.  The JFIF-in-TIFF is the old
way as defined in TIFF6.  It sucked because it broke TIFF's own rule of
not allowing overlapping tag data.  I can see adding decode support for
it, but we definitely shouldn't encode it.

-Justin