2.3 KiB
LUT Extractor
This is a relatively simple command line tool to help get transfer function LUTs and color space chromaticities out of other color processing tools, as long as they support reading and writing OpenEXR image files.
Extracting transfer functions.
First, generate the test image:
lut_extractor --test_image
Load the resulting lut_extractor_2560x1440.exr image into your color processing software and run it through the processing that would convert from encoded color to linear color, and save the processed image as a separate OpenEXR image.
Important notes:
- The new image must be exactly the same resolution as the test image. Any image scaling, blurring, etc. will break this process.
- Ensure that no other color processing is being done other than linearizing the RGB values. For example, ensure that no gamut conversion is being done at the same time.
- Ideally, you should save it as a 32-bit floating point, lossless OpenEXR image. Half float will reduce precision, and lossy compression can cause artifacts in the computed LUT.
Next, run the processed OpenEXR image through lut_extractor like this:
lut_extractor -i filename.exr
This will produce two LUT files, one in .cube format and one in .spi1d format. It will also print Rust code for an attempted analytic fit of the LUT. The printed error percentages only apply to the Rust code, not to the LUT. The LUT is always accurate as long as you followed the steps correctly.
And that's it!
Extracting chromaticity coordinates.
This is essentially the same process as above, except doing a color gamut conversion instead of a linearizing conversion:
Process the test image into CIE XYZ color space, and save it as a separate OpenEXR image. Ensure that only gamut conversion is being done. For example, no non-linear -> linear conversion.
Next, run the processed OpenEXR image through lut_extractor like this:
lut_extractor --chroma filename.exr
It will then print the chromaticity coordinates.
License
This software is licensed under either of
- Apache License, Version 2.0, (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
- MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)
at your option.