ZoneMinder Forums

Good day, apologies for the delay.
currently moving house, so my setup is in pieces
As with the "standard" zmes, i'd like to be able to pull camera, object (person, dog), confidence level, face (recognition), confidence.

Thank you

@tsp84 Good day. Are you passing mqtt to homeassistant to send HA notifications? if yes can you share your automation? I was using the below with the oringinal zm and the json was parsing fine, but it's not doing the job with neozm - id: notify_person_detected alias: Notify person detected # initial ...

Thank you for all your support. I have replaced pyzm and it now detects face with frame_strategy: first

yessss !!! it works..... and following the model_sequence : object, face but frame_strategy: most needs to be like this. if set to first it doesnt work root@UNRAID:/var/lib/zmeventnotification# sudo -u www-data /var/lib/zmeventnotification/bin/zm_detect.py --config /etc/zm/objectconfig.yml --debug ...

ok, made the changes.
also noticed that objectconfig.yml has frame_strategy: '{{frame_strategy}}' but it's not defined like in mlapi.

no detection. this is mlapi.log
https://pastebin.com/5KmfHh9C

Ok, so that fixed it. no errors. Sorry about that, but i couldn't work out that the pushover user/pass is used for the custom script as well. root@UNRAID:/config/hook# sudo -u www-data /var/lib/zmeventnotification/bin/zm_detect.py --config /etc/zm/objectconfig.yml --debug --eventid 140 encrypted ...

i found the error in my gotify config. I've moved it several times so I missed something.

i dont have a 'view_only_username' , I only the default admin and juan

the view_only user is on your config screenshot. maybe it's hardcoded somewhere?

I've given that user full privileges (up to edit where applicable). So it should get the token

stll error at the end root@UNRAID:/var/lib/zmeventnotification# sudo -u www-data /var/lib/zmeventnotification/bin/zm_detect.py --config /etc/zm/objectconfig.yml --debug --eventid 314 encrypted_data = {'gAAAAABhmOO-rh5ZSDmeG_Eu7AhoDpakEVrsli0_vLR4K3N_WpZK5Zl6wGn6lfRdnG8hvMzt8AN-pyoSzI4CA70rC-A134kyjg ...

it's throwing this error root@UNRAID:/var/lib/zmeventnotification# sudo -u www-data /var/lib/zmeventnotification/bin/zm_detect.py --config /etc/zm/objectconfig.yml --debug --eventid 228 encrypted_data ...

thank you. that works and it completed training. now to activate face recognition, i tried (in objectconfig and mlapi) model_sequence : object, face and in face face: general: face_detection_pattern: '{{face_detection_pattern}}' # combine results below same_model_sequence_strategy: 'union' sequence ...

Apologies. Still some errors root@UNRAID:/config/hook# python3 ./zm_train_faces.py 11/20/21 13:31:37.624475 zmes_face_train[4644] DBG1 new_yaml.py:242->[config:init: the supplied config file exists -> '/etc/zm/objectconfig.yml'] 11/20/21 13:31:37.712563 zmes_face_train[4644] DBG1 new_yaml.py:265 ...

Thank you. now this

Good day trying to get face recognition running it throughs the following error: root@UNRAID:/config/hook# python3 ./zm_train_faces.py Traceback (most recent call last): File "./zm_train_faces.py", line 32, in <module> zmesc = proc_conf(args=args, conf_globals=g) TypeError: process_config() missing ...

yes I see it. thank you.
I'll build nvidia and give a test.