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Outdated suggestions cannot be applied.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.
All these changes are for Immich.
Change Made: Deducted a point for the web app.
Justification: The web app on my galaxy S22 partially crops the header. Tested on both chrome and samsung internet.
Change Made: Plus one point on duplicate detection
Justifcation: Immich actually uses CLIP-based AI models to compare images for deduplicating other than just exact duplicates. But I just added one point because of the images it keeps; it simply keeps the largest file (instead of further analyzing which image is more blurry/better quality).
Change Made: Plus three points on face recognition.
Justification: Not only did I find that, at least for my photos, Immich did just as well as photoprism (for its largest model), but also, Immich lets you choose which facial recognition model (big, medium, small). On top of this, you can also adjust the thresholds manually for your image set in the default settings don't work well (which, in my experience, they do). With this added configurability and finding that Immich was handling images of a person in many different types of scenarios very, very well, Immich should be rated AT LEAST as high as photoprism.