home-assistant.github.io/source/_components/image_processing.facebox_face_detect.markdown

layout	title	description	date	sidebar	comments	sharing	footer	logo	ha_category	featured	ha_release
page	Facebox Face Detect	Count the number of faces in an image with Facebox.	2018-05-03 00:00	true	false	true	true	machine-box-alpha.png	Image Processing	false	0.69

The facebox_face_detect image processing component allows you to detect (i.e. count the number of) faces in a camera image using Facebox. The state of the entity is the number of faces detected. Facebox runs in a Docker container, and it is recommended that you run this container on a machine with a minimum of 2 GB RAM. To minimise the time for Facebox to detect faces in an image using this component, on your machine with Docker run Facebox with:

MB_KEY="INSERT-YOUR-KEY-HERE"

sudo docker run --name=facebox --restart=always 8080:8080 -e "MB_KEY=$MB_KEY" -e "MB_FACEBOX_DISABLE_RECOGNITION=true" machinebox/facebox

To enable this component in your installation, add the following to your configuration.yaml file:

{% raw %}

# Example configuration.yaml entry
image_processing:
  - platform: facebox_face_detect
    ip_address: 192.168.0.1
    port: 8080
    source:
      - entity_id: camera.local_file
        name: my_custom_name

{% endraw %}

{% configuration %} ip_address: description: The IP address of your machine hosting Facebox. required: true type: string port: description: The port which Facebox is exposed on. required: true type: string source: description: The list of image sources. required: true type: array entity_id: description: A camera entity id to get picture from. required: true type: string name: description: This parameter allows you to override the name of your image_processing entity. required: false type: string {% endconfiguration %}