facial recognition is terrible
HI,
I have been having difficulties with facial recognition that is doing a terrible job with faces.
It's either picking up a lot of pictures where there are aboslutely no faces, or missing a lot of faces
I followed this guide ; https://immich.app/docs/guides/better-facial-clusters/
But still can't found a setting where it doesn't make up faces and recognizes most of them. I'm using the antelopev2 model
Immich V 1.141.1.
I have tried launching the missing faces job, it won't change anything.
i'm doing a face detection and facial recognition reset each time I change a setting
does anybody have advices on that ?
please@ping
14 Replies
:wave: Hey @Bruce Wayne,
Thanks for reaching out to us. Please carefully read this message and follow the recommended actions. This will help us be more effective in our support effort and leave more time for building Immich :immich:.
References
- Container Logs:
docker compose logs docs
- Container Status: docker ps -a docs
- Reverse Proxy: https://immich.app/docs/administration/reverse-proxy
- Code Formatting https://support.discord.com/hc/en-us/articles/210298617-Markdown-Text-101-Chat-Formatting-Bold-Italic-Underline#h_01GY0DAKGXDEHE263BCAYEGFJA
Checklist
I have...
1. :ballot_box_with_check: verified I'm on the latest release(note that mobile app releases may take some time).
2. :ballot_box_with_check: read applicable release notes.
3. :ballot_box_with_check: reviewed the FAQs for known issues.
4. :ballot_box_with_check: reviewed Github for known issues.
5. :ballot_box_with_check: tried accessing Immich via local ip (without a custom reverse proxy).
6. :ballot_box_with_check: uploaded the relevant information (see below).
7. :ballot_box_with_check: tried an incognito window, disabled extensions, cleared mobile app cache, logged out and back in, different browsers, etc. as applicable
(an item can be marked as "complete" by reacting with the appropriate number)
Information
In order to be able to effectively help you, we need you to provide clear information to show what the problem is. The exact details needed vary per case, but here is a list of things to consider:
- Your docker-compose.yml and .env files.
- Logs from all the containers and their status (see above).
- All the troubleshooting steps you've tried so far.
- Any recent changes you've made to Immich or your system.
- Details about your system (both software/OS and hardware).
- Details about your storage (filesystems, type of disks, output of commands like fdisk -l and df -h).
- The version of the Immich server, mobile app, and other relevant pieces.
- Any other information that you think might be relevant.
Please paste files and logs with proper code formatting, and especially avoid blurry screenshots.
Without the right information we can't work out what the problem is. Help us help you ;)
If this ticket can be closed you can use the /close command, and re-open it later if needed.
Successfully submitted, a tag has been added to inform contributors. :white_check_mark:
Settings I'm using
Why did you pick antelope? Buffalo_l works perfectly for me with default settings
The explanation says the models are stored from biggest to lowest, biggest being better ?
But I'll try the buffalo_l
Hmm that might be true, but I remember reading a long time ago that antelope is just if you wanna experiment;
https://github.com/immich-app/immich/discussions/7838#discussioncomment-8740778
[Discussion] Which of the two facial recognition models, buffalo_l and antelopev2, should I choose? (immich-app/immich#7838)
I tweaked my distance up to 0.7 personally. It does depend a lot on what type of pictures
But if you made changes I’d encourage either refreshing or resetting the face job. Changing the model definitely needs that
Thank you, read that
Does anyone have the default settings ? I don't want to reset everything
I don’t recall the default. Heck, mine might be default
My detection zone is 0.7, recognition distance is 0.5, minimum faces is 5, and the model is buffalo_l
Thank you, I will try those settings.
does this happen to anyone else ???
that's the cat on the pizza even worse ...
what I don't understand is that cat was NOT there previously when I had detection zone to 0.5, and now it's picked up when I increased the detection threshold to 0.6. It's supposed to detect less faces when going up how is it behaving like that ?
