Skip to content
Nabrio

Fall and man-down detection

Nara detects when a person falls or stays lying down for longer than a set time, and sends an urgent alert with the location and an image. It helps protect lone workers, elderly residents, hotel guests and people in public spaces.

Ready to configureNara
The same camera view before detection
camera · raw
A person lying on the grass detected by Nara. Move the lens to compare the raw camera view with what Nara detects.

[ 01 ]The problem

A person who has collapsed or fallen, especially when alone, may not be found quickly.

  1. 01inputCamera over the area
  2. 02processFall detection (pose action model) and lying-down detection held for longer than a set time
  3. 03actionUrgent alert with the location and an image

How it works

Nara estimates each person’s body pose and runs a fall-recognition model on it. A timer confirms that the person is still lying down after a set time, which filters out people who get straight back up. The alert goes out with the camera location and an image.

Works with Nara’s built-in models. You set up the zones, rules and outputs.

What you need

  • Cameras covering the area
  • A PC, server or NVIDIA Jetson running Nara

Nara building blocks

What it improves

  • Response time
  • Lone-worker safety
  • Incident records

FAQ

Questions, answered.

Does it identify people?

No. Fall detection works from body pose, not from faces.

Does any video leave the site?

No. Nara processes video on your own hardware. Data is sent elsewhere only if you build a flow that sends it, for example an alert to your own MQTT broker.

Talk to us about fall detection.

Tell us about your site, cameras and machines, and we’ll show you how it runs on your own hardware, or download Nara and try it yourself.