-
Notifications
You must be signed in to change notification settings - Fork 115
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Support JPEG XL compression in DNG #516
Comments
Good hint. Disappointed that they only specified it for LinearRaw but not for CFA. |
Check the new ColumnInterLeaveFactor as well. |
And now also looks like it's in Samsung Galaxy enhanced ExpertRAW DNGs. |
The stated purpose here describes a nonstandard/proprietary method of encoding raw sensor data I remember seeing a few years ago, but can't find at the moment (and in the process, ALMOST makes it standardized except for one issue, see my last comment) - Take the individual color planes from the CFA, and store them as adjacent images in a monochrome JPEG. A potential use case for this would be hardware that was severely IO-bound but had a reasonably performing hardware JPEG engine (I can think of one specific camera that would REALLY benefit from this, as it has a hardware JPEG engine but USB 2.0 and stores bracketed shots from 25 10-bit sensors, which wouldn't lose too much with nonlinear encoding to 8 bits/sample...), or to allow lossy storage of CFA data with JXL. It seems that JXL + the new RowInterleaveFactor and ColumnInterleaveFactor would support exactly this, but wouldn't that technically require PhotometricInterpretation to be 32803 (CFA)? e.g. "so close yet so far" here. |
Looks like iPhone 16 started using it as well. |
Just added in DNG 1.7, and is allowed for e.g. 3-channel LinearRaw photometric interpretation.
Available (C) implementations:
https://github.com/libjxl/libjxl (reference)
https://github.com/lifthrasiir/j40 (single header, public domain)
The text was updated successfully, but these errors were encountered: