Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is my PR for laxamentumtech/audnexus#530
Like I mentioned in this comment, The centering actually has nothing to do with any sort of facial recognition, it just turned out that the photos in portrait layout look better when they're cropped to the top. So instead of making a new custom option, I just set things up to crop each portrait image to a square centered at the top, and crop each landscape image at the horizontal center.
There is no real need for cropping the horizontal images, as Plex will do the same thing automatically, but I added it anyway in case you'd prefer images to be consistently coming back as square. If you don't want them cropped, I (or you) can just remove L328-L336.
Let me know your thoughts!
Here is what my Authors looked like after running this new agent:
Compared to what they looked like before this change: