# double-take

> Double Take is a self-hosted, open-source web app that unifies facial recognition across multiple detectors and NVRs like Frigate.

- **Category:** Computer Vision
- **Licence:** MIT
- **Latest version:** v1.13.11.8
- **Open source:** oss
- **Languages:** JavaScript, Vue
- **Platforms:** Docker, Docker Compose
- **Tags:** Mqtt, Face Recognition, Home Assistant, Nvr, Frigate, Docker, AI, Computer vision, Facial Recognition, Compreface, Rekognition, Deepstack

## Description

Double Take provides a single web interface and API for facial recognition, connecting to multiple detection services including CompreFace, Amazon Rekognition, DeepStack, CodeProject.AI Server, and Facebox. It integrates with NVRs such as Frigate and Agent DVR by subscribing to MQTT topics and processing camera images automatically. The responsive UI lets users manage matches, train subjects, and configure detectors. A REST API allows external applications to invoke recognition. Results publish to MQTT, with notifications via Gotify or Telegram. The app runs as a Docker image on amd64 and arm64, storing data in a mounted volume. Optional password authentication protects the UI and API. Licensed under MIT.

## Links

- [Homepage](https://github.com/skrashevich/double-take)
- [Website](https://hub.docker.com/r/skrashevich/double-take)
- [Docker](skrashevich/double-take)
- [GitHub Repository](https://github.com/skrashevich/double-take)
- [Docker Hub](https://hub.docker.com/r/skrashevich/double-take)
- [Discord Community](https://discord.gg/3pumsskdN5)
- [Telegram Chat](https://t.me/FrigateNVR)
- [Home Assistant Add-ons](https://github.com/skrashevich/hassio-addons)

---
Source: https://selfhost.directory/project/double-take