> For the complete documentation index, see [llms.txt](https://docs.eyepop.ai/developer-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.eyepop.ai/developer-documentation/deploying/vms/network-optix/running-detection.md).

# Running Detection

Turn EyePop on for a camera and see bounding boxes

With EyePop Vision AI approved and the API key set, you run detection by turning it on for a camera.

{% stepper %}
{% step %}

#### Turn EyePop on for a camera

In the NX desktop client, open the camera's *Camera Settings* > *Integrations*. Turn on *EyePop Vision AI*, choose a **Workflow** (`default` to start), and click *Apply*.
{% endstep %}

{% step %}

#### See bounding boxes

Open the camera's live video. Each detected object gets a box labeled with its track number, class, and confidence, for example `#3 person 91%`. The same class always gets the same color.

The first boxes can take about half a minute to appear while EyePop starts on the camera.
{% endstep %}
{% endstepper %}

### Default behavior

No configuration is required for basic detection. Every camera uses the built-in **`default`** workflow unless you choose another one. It detects common objects with EyePop's `eyepop.common-objects` ability, keeps detections with at least 80% confidence, and tracks each object across frames. To change what a camera detects, see [Workflows](https://docs.eyepop.ai/developer-documentation/deploying/vms/network-optix/workflows.md).

To stop detection on a camera, turn *EyePop Vision AI* off in its *Integrations* settings.

### Next steps

* [Workflows](https://docs.eyepop.ai/developer-documentation/deploying/vms/network-optix/workflows.md) — control what each camera detects.
