[Feature] Auto Image Rotation Detection

[Tyree]

I have searched the existing feature requests to make sure this is not a duplicate request.

• Yes

The feature

Certain apps* have a feature that detects rotated images and curates them for the user to choose to rotate them as suggested or not. Similar to how Immich now detects duplicates, I'd love to see it be able to automagically find these rotated images!
Thanks!

#1695 is similar, but it's for a manual image rotation tool and not for auto detection. Perhaps it could just be extended to include an auto-detect functionality.

Platform

• Server
• Web
• Mobile

[NicholasFlamy]

A message with a suggestion for this, including a link to an existing AI model for this:
https://discord.com/channels/979116623879368755/1129456657995403294/1246841616191324160

Link to existing AI model:
https://github.com/pidahbus/deep-image-orientation-angle-detection

[hexxone]

+1 for the idea
Thanks for reposting this. Would be very cool to see!

[Chuckame]

If we don't want to touch the original photo, we could generate related xmp sidecar using the tiff:Orientation property coming from Adobe xmp specs

[NicholasFlamy]

Pretty sure external libraries can already write sidecar XMP files with image editing and rotating if you have read-only disabled.

[dionorgua]

It looks like Immich doesn't support tiff:Orientation.
It doesn't work for me:
#15549
#15550

[parsapoorsh]

i had this problem, but i couldn't find a lightweight model to do rotation detection, so i glued one together:
https://github.com/parsapoorsh/resnet-ixion

it's size is about ~220 MB, and it has a great performance for images that have an object or face in them (like the images in demo instance)

but it can't detect rotation of scanned papers or an empty wall and stuff like that

[NicholasFlamy]

Holy sh*t I got ROCm working.

[parsapoorsh]

We'd need a check to handle cases like this where it's a logo that seems fine in any orientation. Something like if all angles are within 15% then do nothing.

i agree, i stumbled upon the following images, and having a diff check of something like 15% or averaging all of scores fixes this issue:

that said, it's not supposed to check for rotation of a LOGO, it's trained to find rotation of normal (not spam) images taken from a camera, which would have recognizable objects in them

[NicholasFlamy]

that said, it's not supposed to check for rotation of a LOGO, it's trained to find rotation of normal (not spam) images taken from a camera, which would have recognizable objects in them

People using immich will likely have other pictures in their libraries, so that's gonna come up. That's why I'm thinking of retraining your trained model with a bunch of documents and stuff.

[parsapoorsh]

retraining your trained model with a bunch of documents and stuff.

checkout these datasets:
https://universe.roboflow.com/browse/documents
https://sg-vilab.github.io/event/warpdoc/

also it's possible to synthetically generate data from these. for example putting them in a "screenshot with a camera" format, and other stuff.

but im also thinking about having multiple orientation detection models, fine-tuned for specific type of images.
for example using an OCR model to detect rotation if the main one fails and gets low score.
this doesn't seem to be computationally expensive, because it only calculates when the main model fails.

[NicholasFlamy]

but im also thinking about having multiple orientation detection models, fine-tuned for specific type of images.

If you overcomplicate it, then it'll be harder to maintain in immich.

for example using an OCR model to detect rotation if the main one fails and gets low score.

OCR is something that seems like a lot of people want in immich so maybe that can somehow be integrated if it's added.