fix: FFmpeg sets zero oiio:BitsPerSample #4885
Conversation
Many codecs don't provide bits_per_raw_sample. The bit depth of the luma channel is the closest equivalent to a single bit depth then, while the chroma channels may be subsampled. Tested on basically all the video files on my computer, they all give a non-zero bit depth now. Signed-off-by: Brecht Van Lommel <[email protected]>
|
In case it changes how you want to do this: It is not necessary to set "oiio:BitsPerSample" unless it differs from what is implied by ImageSpec.format. For example, if the standard type you are returning is uint16, but you want to say, "I'm conveying this data via 16 bit ints, and of course 65535 is interpreted as 1.0, but just you know, in the file, it was stored only 10 bits," then you set "oiio:BitsPerSample" to 10. |
|
In my tests this changes a bunch of cases from 0 bits to 8, 10 or 12 bits. So I think it is useful. I could not set it if it matches the format, but other file formats seem to set it unconditionally. |
If that's what all the others are doing, I think it's fine. |
…dation#4885) Many codecs don't provide `bits_per_raw_sample`. The bit depth of the luma channel is the closest equivalent to a single bit depth then, while the chroma channels may be subsampled. Tested on basically all the video files on my computer, they all give a non-zero bit depth that looks correct now. Signed-off-by: Brecht Van Lommel <[email protected]> Signed-off-by: Zach Lewis <[email protected]>
…dation#4885) Many codecs don't provide `bits_per_raw_sample`. The bit depth of the luma channel is the closest equivalent to a single bit depth then, while the chroma channels may be subsampled. Tested on basically all the video files on my computer, they all give a non-zero bit depth that looks correct now. Signed-off-by: Brecht Van Lommel <[email protected]>
Description
Many codecs don't provide
bits_per_raw_sample. The bit depth of the luma channel is the closest equivalent to a single bit depth then, while the chroma channels may be subsampled.Tested on basically all the video files on my computer, they all give a non-zero bit depth that looks correct now.
Checklist:
need to update the documentation, for example if this is a bug fix that
doesn't change the API.)
(adding new test cases if necessary).
corresponding Python bindings (and if altering ImageBufAlgo functions, also
exposed the new functionality as oiiotool options).
already run clang-format before submitting, I definitely will look at the CI
test that runs clang-format and fix anything that it highlights as being
nonconforming.