Repolode
Submit
roflcoopter

roflcoopter/viseron

Self-hosted, local-only NVR and AI computer vision software for object, motion and face detection on your own cameras.

Open-source alternative to:Ring

View on GitHub
Stars
3,652
Forks
453
Open issues
52
Last commit
18 hours ago

What is it?

Viseron is a self-hosted network video recorder (NVR) that runs entirely on your own hardware. It records video from IP cameras over RTSP and applies computer vision to the footage, so you can monitor a home, office or other site without sending video to a cloud service.

The system is built from components that you enable in configuration. Available components cover object detection, motion detection and face recognition, and the project provides a Component Explorer to browse them. Setup is done with Docker, and configuration is edited through a built-in web interface.

The project is licensed under MIT and is actively developed, with community support available through Discord and GitHub issues.

Key features

  • 01Runs fully local, with no cloud dependency for recording or analysis.
  • 02Records from IP cameras over RTSP.
  • 03Provides object detection, motion detection and face recognition as enable-able components.
  • 04Configurable through a built-in web interface.
  • 05Deployed as a Docker container.

Use cases

  • Monitoring a home with existing IP cameras while keeping footage on local storage.
  • Watching an office or shop with object detection alerts instead of cloud-hosted video.
  • Experimenting with computer vision models on your own camera feeds.

Quick start

  1. 1

    Spin up a Docker container as described in the documentation

  2. 2

    Edit the configuration file using the built-in web interface

Things to know

  • Setup relies on Docker and a configuration file, so it takes some self-hosting effort.
  • Getting help is directed mainly to the Discord community rather than a support channel.

In collections

Self-hosted appsComputer vision libraries & tools

Topics

#nvr#computer-vision#object-detection#face-recognition#self-hosted#rtsp#surveillance#docker#coral#cuda

FAQ

What is viseron?
Self-hosted, local-only NVR and AI computer vision software for object, motion and face detection on your own cameras. Viseron is a self-hosted network video recorder (NVR) that runs entirely on your own hardware. It records video from IP cameras over RTSP and applies computer vision to the footage, so you can monitor a home, office or other site without sending video to a cloud service.
Is viseron free to use? What license does it have?
viseron is open source under the MIT license. Check the license terms before using it in commercial products.
Is viseron still maintained?
Actively maintained. The last commit to viseron was on Oct 23, 2026.
What can I use viseron for?
Monitoring a home with existing IP cameras while keeping footage on local storage. Watching an office or shop with object detection alerts instead of cloud-hosted video. Experimenting with computer vision models on your own camera feeds.
How do I install viseron?
Spin up a Docker container as described in the documentation Edit the configuration file using the built-in web interface
Where can I get viseron?
The source code and installation instructions are on GitHub: https://github.com/roflcoopter/viseron

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects