[Feature]: Facial Recognition #1692
Replies: 19 comments 6 replies
-
|
Hey, this is certainly something we have in the plans, however it's very difficult to do it in the way Google photos does it. We want to have this for the first official release of Immich but we want to do it right, so it might not happen for a little while as we investigate the best way to make this happen. |
Beta Was this translation helpful? Give feedback.
-
|
Hey there, Update: |
Beta Was this translation helpful? Give feedback.
-
|
Linking the corresponding PR: #641 |
Beta Was this translation helpful? Give feedback.
-
|
I take a look of Photoprism's code and find they're using pigo to detect faces. Even though pigo is written in Go but it supports Webassembly, maybe it could also be used by Immich? |
Beta Was this translation helpful? Give feedback.
-
|
@xxyzz , @LackesLab was planning to work on Face Recognition feature with state-of-the-art recognition methodology. We will take a look at this once we jump to this feature. |
Beta Was this translation helpful? Give feedback.
-
|
Hey all, sorry for not getting back to you in so long. I've had a lot on my plate and am still working on the feature. The creation of the face patterns is already done. I am currently still developing the way the faces are organized. I hope to find some time in the next weeks. |
Beta Was this translation helpful? Give feedback.
-
No problem! Thank you for letting us know. Again, there is no rush on your part |
Beta Was this translation helpful? Give feedback.
-
|
Hello why not integrate https://github.com/exadel-inc/CompreFace in immich ? |
Beta Was this translation helpful? Give feedback.
-
|
Adding search keywords: Regonize people from pictures |
Beta Was this translation helpful? Give feedback.
-
|
As you consider face tagging, please consider making the metadata readable and writable from the pictures and not just the DB inside Immich. There are many photo management solutions in place today and it seems like almost every one tries to use AI/ML to CREATE metadata, but they most simply store it in their database and not inside the EXIF information inside the photo. Hence when users move between photo management applications the metadata is lost. Given the time horizon people keep pictures (many many years) I would submit that keeping the metadata INSIDE the pictures (or in XMP sidecar) is critical. Think of how people write dates or names on the back of physical paper photos. There seems to be a standard emerging in the industry for face metadata. Digikam is an example of a desktop app that is able to create/read/write/update face metadata. Other image galleries like PiGallery2 or MyLio are able to read and leverage that metadata written by Digikam. There is also a great tool for reading and writing the metadata in exiftool. Some people like myself have spent significant amounts of time tagging photos with faces, keyboards, geolocation and time, and tools that don't read that metadata and/or let us update that metadata going forward are a non-starter. For me, reading existing face metadata is more critical than creating new face metadata with AI/ML. There are other tools already that can create the face metadata (like Digikam). Any new metadata that is not stored with the picture will likely be lost when I inevitably try our a new photo application in the future. |
Beta Was this translation helpful? Give feedback.
-
|
I totally agree with @rhatguy and I was about to comment the same thing. I hope Immich can live together with other DAM like DigiKam 👍 |
Beta Was this translation helpful? Give feedback.
-
|
To extend to this, iOS photos also has face tagging and this data is stored in the EXIF as a collection of bounding boxes. However, the ID it assigns these boxes is unique to each photo (i.e. starts at 0, box 0 in photo X wont be the same face that is in box 0 in photo Y). I assume Apple then has a separate lookup somewhere of 2 possibilities:
Just some thoughts. |
Beta Was this translation helpful? Give feedback.
-
|
I've seen this issue has been marked as Done. Does that mean facial recognition is fully implemented as of now? |
Beta Was this translation helpful? Give feedback.
-
|
I think the issue was only closed to continue the conversation here. GitHub Automation then automatically marked is as done which is wrong. |
Beta Was this translation helpful? Give feedback.
-
|
Re: the feature to read existing metadata, I started some work here to be able to discover the So, for instance, if you've been using something like Digikam to curate face metadata or tags in your photo library, these fields would now be searchable in Immich. |
Beta Was this translation helpful? Give feedback.
-
|
I do like the feature of auto-tagging known faces when it has a high certainty, then ask the user to confirm faces when it is less certain. And a menu to list people, known or unknown, where each person is listed with their auto-cropped face photo, and link to all photos that they've been recognized in. Also +1'd to storing metadata in the photo. |
Beta Was this translation helpful? Give feedback.
-
|
Implemented in #2180 |
Beta Was this translation helpful? Give feedback.
-
|
Is there intention of enhacing what was implemented to read (or write) existing face tags from images (or xmp files), or is the intention for the data to stay within the immich database? |
Beta Was this translation helpful? Give feedback.
-
|
There is an existing PR open to enable sidecar xmp file reading. I think after that is merged there is a possibility that we could add some sort of job to write data back. |
Beta Was this translation helpful? Give feedback.
-
|
Got it. #2466 didn't seem to mention face metadata in the issue, and #2180 didn't mention xmp, so I wasn't sure if/how the two were related. Reading will be a big step forward for those who have spent lots of time tagging existing photots as well as for interoperability between Immich and other apps. Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
Yeah, we generally have this approach of not ever wanting to update the original file. It messes up the hash and makes everything a lot more complicated to manage. Sidecar files seems like a good compromise though. We have not specifically talked about or discussed writing anything (like objects, tags, description, or people) back as part of any PR yet. I think the door is open for that discussion though, especially given the imminent support for (reading) sidecar files. |
Beta Was this translation helpful? Give feedback.
-
|
Hi, I'm just setting immich up and can't work out how to add faces that are unamed to an existing named face that it doesn't think is the same person. How do I do that? If I give it the same name it doesn't merge. Any help would be appreciated. |
Beta Was this translation helpful? Give feedback.
-
|
You can take a look at the video here 😄 |
Beta Was this translation helpful? Give feedback.
-
|
Hi, not 100% sure so asking for confirmation - Do we read face information from the original file? (e.g. digikam allows writing face tags as xmp data to the jpeg file itself and my files have that info) |
Beta Was this translation helpful? Give feedback.
-
|
Hello 👋. I have a mid/large collection of pictures with faces already tagged by an external tool (digikam); I invested a considerable amount of time keep it clean and I'd like to keep my current process. I agree having immich writing faces metadata would be a nice feature; but right know a killer feature for me is to be able to read them from the image metadata/sidecar. I've opened PR #6455 with my initial implementation. The feature is disabled by default; the user must enable it from settings. It's a WIP but hope that fits the expectations from the users (including myself!). Some users that mention this feature in this discussion that might be interested: @rhatguy, @atlas-shrugged08, @cycneuramus NOTE: I developed this patch right on top of v1.92.1 and since then some things have changed on the code that breaks a clean merge. I'm adapting it to the new structure.In the mean time you can take a look or test it yourself; here you have the working branch: https://github.com/bugfest/immich/tree/v1.92.1-metadata-faces EDIT: updates on getting the PR ready to merge |
Beta Was this translation helpful? Give feedback.
-
Feature detail
Google Photos and iPhone Photos have a feature where photos are organized based on faces found in the image.
Also, there are some self hosted solution providers which also have such feature.
It would be great if you can provide such feature in immich.
Platform
Mobile App
Beta Was this translation helpful? Give feedback.
All reactions