I am building a Raspberry Pi - based digital picture frame, using a 4K monitor (TV). The monitor can be adjusted, but not really calibrated (you can’t load in correction curves). The Raspberry Pi can’t handle ICC monitor profiles - it can only drive raw RGB (at least, I haven’t been able to find a way to do this yet).
However - I have a monitor calibration device (a ColorMunki), and the ability to drive the monitor with my MacBook, so I should be able to generate an ICC profile for it.
What I want to do is to develop a Photoshop workflow which uses that ICC profile to “pre-correct” the image, so that when it is displayed from the Pi to the monitor, the colors look correct. Does anyone have an idea on how to do this? I know that Photoshop can use printer profiles to do “soft proofing” on the screen, but I’m not sure that you can save an image with those curves “baked in,” and I’m also not sure if that works with monitor profiles.
You want to convert the image from the camera working space to the working space of the TV you have, empirically determined, and save it that way.
Uggh. It’s far simpler to start with a pre-calibrated display that you put into sRGB mode. This color profile nonsense has never worked because most apps ignore it because it’s a ton simpler to just write your code to handle 3 8-bit numbers.
It also makes far more sense for displays to do all the lifting here : they are logically responsible for their own analog image, and you could even embed sensors in the display so that they self correct themselves as the display ages.