Apple Photos Workflow suggestions

[Adam Barhan]

Pretty new to the app but long time photo nut so excuse any ignorance.

What I am dealing with-
Apple iMac/MacBook Pro computers
Sony A7RIV with ARW RAW files
Apple Photos
iPad Pro
Pixelmator Pro

Can someone showcase a workflow to get the files from the SD card into Apple Photos utilizing the app? I have been importing them directly to Apple Photos and the RAW 123MB files show up but the limits of Apple Photos post processing are apparent.

I would imagine there is a way to achieve the following just can't wrap my mind around it-
RAW/ARW originals safe and backed up in Apple photos
A "layer" of edits that displays in Apple photos, synced across the library
Able to process in a batch for a large group of photos
All meta data/dates/locations in place

I have found the extension section lets me accomplish all of this less the batch section. I would love to be able to batch "ML Enhance" a group of photos from vacation and then go after them individual if needed.

Sorry if this has been covered but I have searched. thanks!

[Andrius]

Hey Adam, you'd be able to do everything, save for the batch editing, which I believe you've discovered yourself. In theory, you might be able to batch edit the RAW files using Automator, converting them to PXD format in the workflow, then find the same images in the Photos app, open them in the extension one by one, press the Command-O keyboard shortcut and find the Pixelmator Files and replace them. But the time saved by using Automator would most likely be lost during this cumbersome import workflow, so you'd be better off clicking the ML Enhance button in each photo in Photos. There aren't really any ways to get around this but we'll keep it in mind for future updates.

[dbourne]

Andrius, where is this ML Enhance button that you speak of? I don't see it in Photos. Do I need to transition to Pixelmator Pro first?

[Andrius]

You'd need to open up the Pixelmator Pro extension in Photos first, yeah. You can find the steps for that here:

http://help.pixelmator.com/pixelmator-pro/1.5.4/#1185

[dbourne]

Thanks. OK, did that. The only way I saw to use Enhanced ML was the key sequence: command-option-U

[dbourne]

Hey Andrius, was that you on the Photo Active podcast this morning? Good job. I learned that ML is more than just enhanced resolution. I'll have to explore the program dome more.

Also, a comment about 'lack' of ML apps. I've dabbled a little with turicreate and ML libraries. A big barrier is gathering the amount of material to learn from. You use millions of photos. I can see that as quite an effort but well-done.

[Andrius]

ML Enhance is also available as a button when the Color Adjustments tool is selected (and using the Shift-Command-M keyboard shortcut). That feature automatically enhances the colors of photos. Option-Command-U is actually for ML Super Resolution, which increases their resolution — is that the option you were looking for?

As for the PhotoActive podcast, yes indeed, that was most certainly me. Hope you enjoyed it! I haven't had a chance to listen to it just yet but I hope I did OK.

Also, a comment about 'lack' of ML apps. I've dabbled a little with turicreate and ML libraries. A big barrier is gathering the amount of material to learn from. You use millions of photos. I can see that as quite an effort but well-done.

Actually, as far as we've found, the challenges with each machine learning project are often different. Sometimes, the hardest thing is gathering the dataset. Other times, it can be difficult to create a good loss function (the thing that measures the accuracy/quality of a model and helps it learn). And sometimes, a good enough model might just be too big and slow to be usable in everyday image editing tasks.

[marciachrist]

Is there a link to the PhotoActive podcast?

[marciachrist]

Found it: https://www.photoactive.co/home/episode-60-pixelmator

I'm going to listen to it now. Thanks, Andrius!

[dbourne]

The ML Super Resolution in the latest update was what I noticed. I didn't realize there were other ML enhancements until I listened to the podcast.