Custom model - one of the best features of Clarifai. If you just want face recognition the OpenCV component works
I was able to get microsoft face working. It is correctly firing automations when I walk in front of the assigned camera. I’ll need to test it now with the whole family and see how accurate the ID is.
One thing that would be neat would be if I can get an automation to fire when there is a face detected that is not known.
28-3-2018 Component is working, I was probably using incorrect config before.
@Ouassimf did you ever get a solution for the quota issue? i have everything working, however with the trial susbscription you can only use the API 30000 times in a month, and i’m spending that amount in a week.
is there any way to only trigger face recognition based on some events??
@peteonus
I realized that anyway I dont want to rely on cloud. Went for dlib instead which does NOT work out of the work on hassio / hassbian.
I installed everything on a rpi3 and modified had to modifiy the dlib component face_recognition requirement from 1.0.0 to 1.2.2 (latest version that DO work with scipy)
I would recommend to use this Face detection with Docker/Machinebox
Works great for me. Thank you @robmarkcole
Just having a go at setting this up… I’ve got the config in for microsoft_face
microsoft_face:
api_key: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
azure_region: westeurope
I’ve restarted hass, but can’t see any of the services available as described in the documentation, as a result I get the following error:
Unable to find service microsoft_face/create_group
Any clues where I’m going wrong?
Cheers
Andy
@mcinnes01 i had microsoft face fully working however didn’t follow the instructions on HASS help, here are my suggestions.
- verify your region, it has to match the region where you created the azure account
- API key, couldn’t get to work with the KEY1 i had to use the KEY2 always
- i didn’t create users, groups or trained the component using the HASS, i’ve used the toolbox provided by microsoft, you can find it on the website, it’s probably the longest way to do it but its the most realible because you would see an instant response.
i’m still dealing with the 30K monthly limitation of the API requests, but is working perfectly.
@armdan it works great indeed, the only problem is you have to train it every time you restarted because the free version wont let you save the state, anyway to overcome this?
I believe the great @robmarkcole trying to work on it. He is trying to make a folder where you put all your pics and it gets recognized automatically. Let’s see and hope but still there is no alternative for me. I tried them all and this one is the best and fastest.
It looks like the create part of the component is broken.
@peteonus What does the toolbox look like? How do you get to it?
Stuck on adding groups via UI and Json errors.
Think keys are working (either Key 1 or 2) only way I can tell is by entering an invalid kep system throws a log error, else now way of telling.
But can’t create group and can’t find anywhere within Azure to create users or groups.
Running bleeding edge of HA @ 86.3
Any pointers?
I was able to get everything working throughout this post but the 2 faces I add it will show the names properly and it appears to have a guid assigned to them each but when I use the camera with 1 person or 2 people it comes back stating it finds it but shows this…
I’ve tried to delete all items and start over, trained multiple times but it always comes back as [Object object] for the name instead of the name of the person.
Hopefully someone has run into this or has a suggestion to work around it.
Hi,
Were you able to find a solution? My face works fine and returns my name. Also recognizes my wife’s face but not returning the name, not even unknown.
The Home assistant addon seems to create a persongroup using recognition_01 (old face recognition model). The API now supports recognition_02
Is there a way to request that recognition_02 is used?
or does the microsoft face home assistant platform need to be updated?
For anyone like me who also came across this issue: adding names of the users directly via the API (e.g. via Postman or the Azure web interface) causes the names not to be retrieved in HA. So it is important that you add the names through the Face Integration. After that, you can also add images by directly communicating with the API (which results in direct responses as mentioned earlier in this threat).
The toolbox is part of the Microsoft Azure Development environment and looks like this
It can be found under ‘Quick Start’ and then the link under point 2, ‘API-Console’
I am trying to add the microsoft face unsuccessfully
I have created this on Azure Cloud but there is an endpoint in the configuration on Azure which was never used…
I have added in the yaml file but I am getting error after I restart HA.
microsoft_face:
api_key: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
azure_region: eastus
Logger: homeassistant.setup
Source: setup.py:162
First occurred: 10:21:48 AM (1 occurrences)
Last logged: 10:21:48 AM
Setup failed for microsoft_face: Integration failed to initialize
Logger: homeassistant.components.microsoft_face
Source: components/microsoft_face/init.py:317
Integration: microsoft_face (documentation, issues)
First occurred: 10:21:48 AM (1 occurrences)
Last logged: 10:21:48 AM
Error 403 microsoft face api https://eastus.api.cognitive.microsoft.com/face/v1.0/persongroups
I’m having the same issue - did you get anywhere with this? Thanks!