---
title: Grouping
slug: enterprise/grouping
docTags: 
createdAt: 2026-04-08T13:07:18.768Z
---

# How it works

The face recognition rule detects recognized people in a scene. 

When an person's face is uploaded to a watch list, any rule with face recognition turned on will generate an event if the recognized person's face from the watch list is detected. 

# How it detects

IRIS+ detects and tracks people in the scene. When the number of people standing
in close proximity to one another within the region of interest reaches the
threshold you set, and they remain grouped for the dwell time, the rule
generates an event with a short clip and bounding boxes around the people
involved.

Grouping is evaluated on people only. Three settings shape it:

- **Number of people** — the minimum group size that should trigger an event.
- **Proximity** — how close together people must be to count as a group (for
  example within roughly one meter of one another).
- **Dwell time** — how long the group must persist before the event is
  generated.

# How to set up

1. Open the camera you want to configure and add a new rule.
2. Select **Grouping** as the rule type.
3. Draw the **region of interest** to monitor.
4. Set the **number of people** that defines a group.
5. Set the **proximity** and **dwell time**.
6. Save and activate the rule. It can run alongside other rules on the same
   camera and be scheduled to active hours.

# What to expect

- An event is generated once the required number of people are grouped closely
  together for the full dwell time.
- Every event includes an event clip and bounding boxes around the people in the
  group.
- Grouping is designed for smaller gatherings and is intended for group sizes up
  to about 10 people. To monitor larger, denser crowds, use the Crowd Density
  rule instead.
- Best results come from clear views where individuals can be separated; in very
  dense scenes the per-person approach becomes less reliable.