Skip to content

FaceStream.AI

Documentation

Overview

18 articles

Getting started

Getting started

Connect your first camera, add a person and check recognition. Then choose a guide for Home Assistant, Loxone or notifications.

Getting started

Installation

Choose the installation steps for Docker, Synology, another NAS or Linux. Start FaceStream, then add your camera. Optional maintenance and advanced settings follow.

Getting started

Editions and licence

What FaceStream.AI Free, Pro, Business and Enterprise include, how to enter a licence, and how activation and the update period work.

Setting up

Adding a camera

Add IP cameras, video doorbells and the Loxone Intercom: ONVIF camera search, RTSP and MJPEG addresses by brand, output size, detection zones.

Setting up

People and photos

Enrol the people FaceStream.AI should recognise: which photos work best, how many you need, and how to rename or delete someone.

Setting up

Recognition and requests

Continuous face recognition or recognition when the doorbell rings: the settings, the liveness check, the events and the trigger API.

Setting up

Connections and rules

Send face recognition events to Home Assistant, Loxone, KNX, MQTT, webhooks or your phone: connections, rules, placeholders, delivery log.

Setting up

Events and visits

The visit log: who was recognised at which camera, when and for how long, with the pictures — and how long they are kept.

Integrations

Face recognition in Home Assistant

Face recognition for Home Assistant via MQTT discovery: a device per camera, automations for arrivals, strangers and the doorbell, a live dashboard.

Integrations

Face recognition with Loxone

Face recognition for the Loxone Intercom and Miniserver: the bell starts the recognition, the name lands on a virtual input. Step by step.

Integrations

KNX

Write a KNX group address when somebody is recognised: a switch, a number or a text over a KNXnet/IP gateway, with examples.

Integrations

MQTT, webhooks and UDP

Connect face recognition to Node-RED, ioBroker, openHAB, n8n or your own scripts over MQTT, HTTP webhooks or UDP.

Integrations

Push messages, e-mail and syslog

A push message with the picture via Telegram, Pushover or ntfy, an e-mail with the face and the whole picture, or a line in your syslog collector.

Use cases

Use case: smart home

Face recognition recipes for the smart home: a doorbell that knows who rang, children home from school, couriers, the driveway at night.

Use cases

Use case: hotels and hospitality

Face recognition in hotels, with consent: greet returning guests who opted in, and hear about unknown people in staff areas after hours.

Use cases

Use case: security and restricted areas

Face recognition for server rooms, entrances after hours and restricted areas: alerts with the picture, e-mail records, syslog.

Help

Troubleshooting

Cameras without a picture, people not recognised, messages that do not arrive, wrong times, a forgotten password — and where to look first.

Help

Privacy and your data

What FaceStream.AI stores, where and for how long, what leaves your network, and how to remove a person entirely.