Dlib Face Identify

The dlib_face_identify image processing integrationIntegrations connect and integrate Home Assistant with your devices, services, and more. [Learn more] allows you to use the Dlib through Home Assistant. This platform allow you to identify persons on camera and fire an event with identify persons.

For using the result inside an automation rule, take a look at the integration page.

Note

This integration is only available on Home Assistant Core installation types. Unfortunately, it cannot be used with Home Assistant OS, Supervised or Container.

Configuration

To enable Dlib Face Identify, add the following lines to your configuration.yamlThe configuration.yaml file is the main configuration file for Home Assistant. It lists the integrations to be loaded and their specific configurations. In some cases, the configuration needs to be edited manually directly in the configuration.yaml file. Most integrations can be configured in the UI. [Learn more] file. After changing the configuration.yamlThe configuration.yaml file is the main configuration file for Home Assistant. It lists the integrations to be loaded and their specific configurations. In some cases, the configuration needs to be edited manually directly in the configuration.yaml file. Most integrations can be configured in the UI. [Learn more] file, restart Home Assistant to apply the changes.

# Example configuration.yaml entry
image_processing:
 - platform: dlib_face_identify
   source:
    - entity_id: camera.door
   faces:
     Jon: /home/hass/jon.jpg
     Bob: /home/hass/bob.jpg

Configuration Variables

source list Required

List of image sources.

entity_id string Required

A camera entity id to get picture from.

name string (Optional)

This parameter allows you to override the name of your image_processing entity.

faces list Required

List of faces sources.

confidence float (Optional, default: 0.6)

How much distance between faces to consider it a match. Using tolerance values lower than 0.6 will make the comparison more strict.

Note

If the platform fails to load because it could not install its requirement, install cmake: sudo apt-get install cmake.