Entities

Entities represent recurring subjects detected across your cameras - people, vehicles, and other recognizable objects. ArgusAI automatically identifies and tracks these subjects over time.

Understanding Entities

What Are Entities?

Entities are persistent records of subjects that appear in multiple events:

• People: Regular visitors, family members, delivery personnel
• Vehicles: Your car, neighbor's vehicles, delivery trucks
• Other: Future support for pets, packages, etc.

How Entities Work

1. AI analyzes an event and extracts subject details
2. System checks for matching existing entities
3. If match found, event is linked to that entity
4. If no match, a new entity may be created
5. Entity history grows with each linked event

Entity Attributes

Attribute	Description	Example
Type	Entity category	Person, Vehicle
Name	Display name (if set)	"White Toyota Camry"
First Seen	When first detected	Dec 20, 2025
Last Seen	Most recent appearance	Dec 25, 2025
Event Count	Number of linked events	42

Vehicle-Specific Attributes

Attribute	Description	Example
Color	Vehicle color	White, Black, Silver
Make	Manufacturer	Toyota, Ford, Tesla
Model	Vehicle model	Camry, F-150, Model 3

Browsing Entities

Entity List

The main Entities page shows all recognized entities:

• Thumbnail: Representative image
• Name: Entity identifier
• Type Badge: Person or Vehicle
• Event Count: Number of appearances
• Last Seen: Most recent timestamp

Filtering

Filter entities using:

Filter	Options
Type	All, Person, Vehicle
Sort By	Last Seen, First Seen, Event Count, Name
Search	Text search in entity names

Entity Cards

Each entity card shows:

• Primary thumbnail from recent event
• Entity name or auto-generated label
• Type indicator (person/vehicle icon)
• Quick stats (events, last seen)

Entity Details

Viewing an Entity

Click any entity to see full details:

Overview Section

• Large representative image
• Entity name and type
• All extracted attributes
• First and last seen dates
• Total event count

Timeline Section

• Chronological list of all linked events
• Event thumbnails and timestamps
• Camera location for each appearance
• Click events to view details

Activity Pattern

• Visual representation of when entity appears
• Time-of-day distribution
• Day-of-week patterns
• Helps identify regular visitors

Editing Entity Details

Customize entity information:

1. Click Edit on the entity detail page
2. Modify:
   • Name: Give a friendly name
   • Notes: Add personal notes
3. Click Save

Managing Event Links

Why Manage Links?

AI matching isn't perfect. You may need to:

• Remove incorrectly linked events
• Add events that should be linked
• Merge duplicate entities

Unlinking Events

Remove an event from an entity:

1. Open the entity detail page
2. Find the incorrect event in the timeline
3. Click the Unlink (X) button
4. Event is removed from entity

The event remains in the system, just unlinked from this entity.

Linking Events

Add an unlinked event to an entity:

1. Open the event you want to link
2. Click Add to Entity
3. Search for the correct entity
4. Select the entity from results
5. Event is linked immediately

Merging Entities

Combine duplicate entities:

1. Go to the Entities page
2. Select entities to merge (checkboxes)
3. Click Merge Selected
4. Choose which entity to keep as primary
5. Confirm the merge

After merging:

• All events from both entities are combined
• The non-primary entity is deleted
• Attributes are merged (primary takes precedence)

caution

Entity merges cannot be undone. Review carefully before confirming.

Entity Alerts

Create notifications for specific entities:

1. Open entity detail page
2. Click Create Alert Rule
3. Configure notification settings:
   • Notification Type: Push, Webhook
   • Schedule: When to alert
   • Cooldown: Minimum time between alerts
4. Save the rule

Now you'll be notified when that specific entity is detected.

Alert Use Cases

• Know when your car leaves/returns
• Track when a specific delivery person arrives
• Monitor for unfamiliar vehicles
• Get notified of repeat visitors

Learning from Corrections

ArgusAI learns from your manual corrections:

How Learning Works

Action	What System Learns
Unlink event	These features don't match this entity
Link event	These features do match this entity
Merge entities	These variations represent the same subject

Improving Accuracy

To improve entity matching over time:

1. Regularly review recent entity assignments
2. Correct any obvious mistakes
3. Merge duplicate entities promptly
4. Provide clear entity names

Tips

Naming Entities

Good names help you identify entities quickly:

Type	Good Names	Avoid
Vehicles	"White Toyota Camry", "Amazon Van"	"Car 1", "Unknown"
People	"Mail Carrier", "Neighbor John"	"Person", "Unknown Person"

Keeping Entities Clean

• Review entities monthly
• Merge obvious duplicates
• Delete obsolete entities
• Archive entities for subjects that no longer appear

Privacy Considerations

• Entity data is stored locally
• No facial recognition is used
• Vehicle matching based on color/make/model
• You control all entity data