Roboflow + OBS

Run a Roboflow model on your OBS Studio output in the browser and switch scenes or move sources when it detects a gesture.

OBS Studio is free, open-source software for recording and live streaming. Our tutorial runs a gesture model on the OBS output and changes the stream in response.

OBS's Virtual Camera exposes the program output as a webcam. A web page picks that camera, runs a Roboflow model in the browser with your publishable key, and connects to OBS's WebSocket server with the obs-websocket-js library. Detections trigger calls such as SetCurrentProgramScene to switch scenes and SetSceneItemTransform to move a source.

How to connect OBS to Roboflow

4 steps
  1. Download the controller

    Download the roboflow-gesture-controller zip linked from our tutorial and open its roboflow.js file.

  2. Enable the OBS WebSocket server

    In OBS, open WebSocket Server Settings, enable the server and note the IP, port and password.

  3. Add your connection and model details

    Enter the ws://IP:PORT address and password in roboflow.js, along with your Roboflow publishable key, model ID and version.

  4. Start the Virtual Camera

    Click Start Virtual Camera in OBS, open index.html in Chrome and select OBS Virtual Camera as the camera.

What teams build with OBS

  • Switch scenes with a hand gesture

    Change the live scene when the model sees a gesture, without touching the keyboard.

  • Move overlays based on what is on screen

    Reposition a source when the model detects something in the frame.

Questions

The tutorial uses OBS's WebSocket server and Virtual Camera, configured in OBS's settings. It connects with the community obs-websocket-js library.

Yes. Replace the model ID and version in roboflow.js with your own and add your publishable key.

Why are companies choosing Roboflow?

The end-to-end platform for vision AI: create datasets, train models and deploy to production.

  • Flexible deployment modes

    Move from a scalable cloud API to edge hardware like NVIDIA Jetson without managing complex dependencies.

  • Integrated workflow builder

    Combine multiple AI models, custom business logic and third-party integrations into one production-ready pipeline in a low-code editor.

  • Model monitoring and alerts

    Dashboards track confidence, inference time and request volume for your deployed models, and email alerts tell your team when confidence suddenly drops.

View all
Get started

Build on the Platform

For developers, engineers, and technical founders who want to get hands on. Try the free tier; the docs are open.

Bring it into your operation

For heads of AI, operations leaders, and enterprise teams. Bring a known problem, or work with us to find the first one worth taking on.