Setting Up Vision AI FaceID for Drivers

  • Updated

How to register driver photos and enable automatic driver identification

 

Overview

Vision AI FaceID automatically identifies who is driving a vehicle — no tags, no log-ins, no driver action required. When a driver gets into a Vision AI-equipped vehicle, the in-cab camera captures their face and matches it against reference photos saved in their Argus profile. Their name then appears on live tracking, event details and in trip history in real time.

This feature is available to Vision AI plan customers only.

 

When to use it

Set up FaceID when your fleet runs shared or pooled vehicles, where more than one driver uses the same vehicle across different shifts and you need accurate driver attribution.

You should also complete this setup whenever a new driver joins your team and will be using a Vision AI-equipped vehicle. Each driver needs at least five reference photos registered in their profile before the system can identify them.
 

How to use it

Before You Start

Make sure the following are in place before registering photos:

  • Your organisation is on the Vision AI plan
  • The driver has an existing User profile in Argus
  • Their vehicle has a Vision AI dashcam fitted and active

Register a Driver's Face Photos

  1. In Argus, go to Settings > Users & Drivers
  2. Open the driver's user profile
  3. Scroll to the Vision AI - FaceID Registration section
  4. Upload 5 photos of the driver from either Facial Database OR Local Photo*
  5. Select Save once your photos have passed quality checks

Vision AI FaceID Registration section in the User profile page
 


Photo Requirements

Each photo is automatically checked for quality when uploaded. The system shows a score from 0 to 100 — only photos scoring above 40 will be accepted. Photos that fail will be highlighted; remove and replace them before saving.

*Local Photo Criteria:

  • Accepted formats: .jpg and .png only
  • Maximum file size: 500 KB per photo
  • Quality score: Must be above 40 out of 100
  • Number of photos: 5 per driver
  • Photos must be at least 300 × 300 pixels
  • Face area min. 100 × 100 pixels
  • The face must be clearly visible, front-facing (≤15°)
  • Ensure even lighting and avoid shadows, glare, or cropped faces
     

Tips for getting a good quality score

  • Use clear, well-lit, front-facing photos
  • Avoid sunglasses, hats, or heavy shadows across the face
  • Uploading 5 varied photos per driver improves recognition accuracy
  • Higher quality scores mean more reliable identification in the vehicle

Example of successful photo upload

Example of failed photo upload

 

How Identification Works

Once a driver's photos are saved, the system handles everything automatically.

  1. The driver gets into the vehicle and starts it
  2. The Vision AI camera captures a photo of the driver
  3. Argus compares the capture against the driver's registered photos
  4. If a match is found, the driver's name is attached to the vehicle's tracking data in real time

The camera checks every 60 seconds while the vehicle is moving. No further action is needed from the driver or fleet manager.


Update or Remove Photos

To change a driver's registered photos at any time:

  1. Go to Settings > Users & Drivers
  2. Open the driver's user profile
  3. In the Vision AI - FaceID Registration section, select the photo you want to remove
  4. Delete the photo and upload a replacement if needed
  5. Select Save
     

Troubleshooting

  • The FaceID Registration section is not visible
    This section is only available on the Vision AI plan. Contact your Argus account manager to find out more.
     
  • My photo file is being rejected
    Vision AI only accepts .jpg and .png files. Re-export or rename your file and try again.
     
  • A driver is not being identified on trips
    Check their registered photos — delete any with a low quality score and upload clearer replacements. 
     
  • The driver shows as unidentified on some trips but not others
    Add more photos in varied lighting conditions to improve identification accuracy.

 

 



Related Articles


 

Was this article helpful?

0 out of 0 found this helpful

Have more questions? Submit a request