Before I get a "lmgtfy" response, I've been searching... lots, but it's tough to find results for a term/classification you're not quite sure of...
I'm wondering if there are any open source server side applications/libraries that operate with a CLI, or library interface, that can manipulate/process images the same way that Google Photos does with their "stylized" creations, preferably with an AI of sorts so that it requires minimal input. There are open source AI's for object and subject identification, so I have to believe that there is work being done on auto-manipulation of images to apply filters, styles, etc.
My goal is to replicate the features (as best as I can) of Google photos on a private "home server". Things like facial recognition, organization, etc. are actually all pretty straightforward, but this is the final piece of the puzzle that would allow me to privatize my images without giving up the benefits of those commercial services.
OpenCV is the closest thing I've found, but I'll be honest, the site, FAQ, tutorials, don't quite clear up what it's actually capable of doing. I see that it does things like blending, stitching, changing colors/levels, etc. but as far as automatically recognizing which images might benefit most from a specific filter or form of processing...??
Are there others that are more in line with this?
EDIT: By open source, I meant to say FOSS (Free and Open Source), not worried if it doesn't do everything under the sun, as long as it has some solid basics covered.
For OS, we're on CentOS8, so anything that'll compile or install via package to it or RHEL/Fedora would be ideal. Libraries that plug into Python, C, etc. would be fine, as long as I can make an exec/command call to the OS, I'm pretty open to any option. Beggars can't be choosers! :)