Custom Vision AI for Verkada Cameras

Add custom AI to the Verkada cameras you already own. Roboflow trains a model on footage from your own sites and runs it on your Verkada video, in the cloud or at the edge, so you can spot spills, watch forklift zones, and check PPE, and send alerts to the tools your team already uses.

An AI-detected spill (94%) on the floor of a warehouse aisle lined with shelves of boxes.

Verkada Cameras AI Vision Solution

Verkada made enterprise security cameras simple: plug in a PoE camera and it shows up in the Command cloud platform with onboard storage and no NVR. Its built-in analytics cover people and vehicles well. When you need to detect something specific to your operation, Roboflow trains a model on your own footage and runs it on the streams your Verkada cameras already produce.

Verkada's Dome, Mini Dome, Bullet, Fisheye, Multisensor, and PTZ series all support low-latency RTSP streaming. For most indoor spaces we recommend a Verkada Dome (opens in a new tab); use a Bullet for long aisles and lots, and a Multisensor to cover a large area with one mount.

  • Featured: Dome, Bullet, Fisheye, Multisensor, and PTZ series
  • Attributes: Hybrid cloud, onboard storage, PoE, no NVR
  • Integration: Send alerts and events to Slack, webhooks, PLCs, MES, and dashboards

How to Connect a Verkada Camera to Roboflow

There are three common ways to get Verkada video into a Roboflow model:

  • Cloud Streaming, nothing to install: Create an API key in Verkada Command under Organization Settings > Verkada API, then add your Verkada organization in Roboflow Cloud Streaming. Your cameras show up as sources, and Cloud Streaming runs your Workflow on managed GPUs.
  • Edge device on site: Turn on RTSP (opens in a new tab) for the camera in Verkada Command (camera settings > Device > Real Time Streaming Protocol, with a username and password), add an RTSP Stream in Roboflow Deployment Manager, paste the camera's RTSP URL, and choose a Workflow. Video stays on your network and Roboflow Inference runs the model on every frame.
  • Python: Point Roboflow Inference at the RTSP URL in a few lines of code and handle each result in your own app.

Tip: On RTSP, each Verkada camera supports up to two RTSP streams at a time, with more on some Multisensor models. Command gives you a standard and a high-resolution URL; start with standard for large objects like people and forklifts, and use high resolution for small details.

Example App: Spill Detection with a Verkada Camera

Here is how a warehouse or store can use the Verkada cameras it already has to catch spills before someone slips:

  • Pick the Verkada Dome and Bullet cameras that look down aisles, near machines, and around loading areas.
  • Pull frames from those cameras, including past incidents, and label spills, leaks, and puddles in Roboflow.
  • Train an RF-DETR (opens in a new tab) segmentation model on those images in Roboflow.
  • Build a Workflow that flags a spill that stays in view for more than a minute and sends an alert to Slack or a webhook with the camera name and the frame.
  • Run it on each camera through Cloud Streaming or an edge device on site.

See leak and spill detection for the full use case and safety zone monitoring for keeping people out of hazard areas. The same cameras can run forklift detection in the same Workflow.

More Applications for Verkada Cameras

Whether you run Verkada cameras at one site or hundreds, you train the model once and run the same Workflow on every stream.

Run vision models on Roboflow hardware

Roboflow Edge computers and the AI1 Smart Camera arrive ready to run your models. Each one ships with RoboflowOS, the operating system we build and maintain, so fixes and security patches reach your devices over the air without a site visit.

Roboflow Edge computers
  • Settings that survive a restart

    RoboflowOS rebuilds static addresses, gateway, DNS and camera-port tuning from a persistent store each time the device starts.

  • Updates over the air

    Security patches, camera driver fixes and NVIDIA platform upgrades come in from the Roboflow platform, separate from your models and Workflows.

  • Tested on the hardware you get

    We build the inference runtime too, so we test every OS image against it on the same boards we ship.

  • Locked down by default

    The OS image is read-only and ships without SSH, a desktop or a local login. Your team reads logs, checks streams and views telemetry in Deployment Manager, the Roboflow tool for managing devices remotely.

  • Five years of updates included

    Every appliance comes with five years of RoboflowOS updates from the purchase date, plus a one-year hardware warranty.

RoboflowOS on NVIDIA Jetson

  • Smart Camera

    AI1

    The AI1 puts a camera, a light and a computer in one housing. It pairs an 8 MP (3840 x 2160) camera and LED ring with an NVIDIA Jetson Orin NX 8GB. Teams use it to prove out an edge vision project on a fixed inspection view, counting parts a few inches away or spotting defects dozens of feet away.

    Memory
    8 GB
    Cameras and networking
    Integrated 8 MP camera and LED ring1 x GbE
    1 stream
  • Roboflow Edge

    RF-NX16

    The RF-NX16 handles stations with several cameras. It has six PoE ports on a shared 150 W budget, two more GbE ports and an NVIDIA Jetson Orin NX 16GB.

    Memory
    16 GB
    Cameras and networking
    6 x GbE PoE with a 150 W total budget2 x GbE
    1–4 streamsRF-NX16 Spec Sheet
  • Roboflow Edge

    RF-AGX64

    The RF-AGX64 runs higher-resolution cameras and bigger models. It has 64 GB of memory on an NVIDIA Jetson AGX Orin, four PoE ports and two 10 GbE ports.

    Memory
    64 GB
    Cameras and networking
    4 x GbE PoE2 x 10 GbE
    4+ streams, line scanRF-AGX64 Spec Sheet
  • Roboflow Edge

    RF-THOR128

    The RF-THOR128 is our most powerful appliance and runs on the NVIDIA Jetson T5000 with 128 GB of memory. It has one 5 GbE port and a 100 GbE QSFP28 port that breaks out to four 25 GbE links, and its cameras connect through a PoE switch.

    Memory
    128 GB
    Cameras and networking
    1 x 5 GbE and 1 x 100 GbE QSFP28Breaks out to 4 x 25 GbE
    8+ streams, edge VLM inferenceRF-THOR128 Spec Sheet

One GbE PoE camera port

Already running Verkada cameras? Connect them to an RF-NX16, RF-AGX64 or RF-THOR128 and manage the whole station from Roboflow.

Not sure which one fits your line?

Send us your cameras, Workflow and line rate, and we’ll create a recommendation for you.

AI for Verkada Cameras Workflows

Use Verkada cameras? Explore Workflows you can get started with to create computer vision apps in minutes. Run your Workflows on Verkada video with Roboflow Cloud Streaming, or with Roboflow Inference on an edge device. Integrate AI by connecting to existing PLC and HMI solutions via standard protocols like MQTT or OPC-UA.

How to Setup Verkada Cameras with AI

Verkada cameras can be used for computer vision tasks such as object detection and tracking. Connect your Verkada organization to Roboflow Cloud Streaming with an API key and run Workflows on managed GPUs, or turn on RTSP and run Roboflow Inference on an edge device on the same network.

  • Meet Roboflow AI1

    AI1 packages onboard AI compute, integrated lighting, and an industrial camera into one plug-and-play vision system.

  • How to Use Your Current Cameras for Vision AI

    Almost any IP camera that supports RTSP, HTTP-MJPEG, ONVIF, or RTMP can feed a computer vision pipeline, meaning teams in manufacturing, retail, or logistics rarely need to replace hardware to start running models on live video.

  • The Vision AI Center of Excellence Blueprint

    A playbook for turning vision AI into a capability your own teams run, with a maturity model, the three CoE roles, and a 10-question diagnostic.

Talk to a Vision AI engineer about your Verkada cameras.

Bring us your Verkada cameras and your toughest monitoring problem, and we'll map a working solution.

Ask us about:

  • Solution architecture that fits your Verkada cameras, network, and sites
  • A live demo on footage from your own Verkada cameras
  • Deployment options: Cloud Streaming on existing cameras, edge, on-prem, or air-gapped
  • ROI modeling against incidents, manual review time, and per-stream pricing across sites

Over 80,000 organizations build with Roboflow.

  • Rivian
  • Pella
  • Chobani
  • USG Corporation
  • BNSF Railway
  • American Woodmark
Start where you are

Additional resources