Troubleshooting Kuna Facial Recognition Not Working
Kuna's facial recognition feature is a powerful tool designed to make your motion alerts smarter and more meaningful. Instead of just getting a generic "Motion Detected" alert, it can tell you when a specific family member has arrived home. However, if it's not working as expected, a few key factors are likely at play.
This guide will help you troubleshoot and resolve the most common issues with Kuna's facial recognition, allowing you to get the most out of this premium feature.
Prerequisite: Check Your Subscription Plan
Before diving into technical troubleshooting, it's essential to confirm that your Kuna account has an active Premium Subscription. Facial recognition and other advanced AI features are not available on the free plan.
- You can check your subscription status within the Kuna app or by logging into your account on the Kuna website. If you are not subscribed, you will need to upgrade to enable this feature.
Step 1: Optimise Camera Placement and Lighting
The performance of any facial recognition system is highly dependent on the quality of the image it receives. The camera needs a clear, well-lit view of a person's face to work effectively.
- Camera Height: Install your Kuna device at a height where it can capture faces, not just the tops of heads. The ideal height is typically around 2 to 2.5 metres (7-8 feet) from the ground.
- Camera Angle: Angle the camera so it points towards where people will naturally be looking as they approach your door. Avoid angling it too high or too low.
- Lighting is Critical: This is the most important factor.
- Daytime: Ensure there isn't a strong backlight (like the sun directly behind the person) that would put the face in shadow.
- Night-time: The area must be well-lit. Relying solely on the camera's night vision is not enough for accurate facial recognition. A good, bright porch light is essential. The system needs to see facial features, not just a ghostly infrared image.
Step 2: Build and Train Your Face Profiles
The system learns over time, but you need to help it.
- Create Profiles: In the Kuna app, navigate to the settings for facial recognition. Here, you can create profiles for individuals you want the system to recognise (e.g., "John," "Sarah," "Postman").
- Assign Detected Faces: When Kuna detects a person but doesn't recognise them, the event will appear in your timeline. You will have the option to assign that detected face to one of your created profiles.
- Be Consistent: The more images of a person you correctly assign to their profile, the more accurate the AI will become at recognising them in the future from different angles and in different lighting conditions. Don't be discouraged if it makes mistakes at first. Diligently correcting and assigning faces is part of the training process.
Step 3: Understand the Limitations
Even with perfect setup, there are situations where facial recognition can struggle.
- Obscured Faces: The system will have difficulty if a person is wearing a large hat, sunglasses, a face mask, or has their head turned away from the camera.
- Fast Movement: If someone walks quickly past the camera without looking at it, there may not be enough time to capture a clear facial image.
- Distance: The person needs to be reasonably close to the camera (typically within 3-5 metres) for their facial features to be clear enough for analysis.
By ensuring your camera is properly positioned, the lighting is adequate, and you actively train the system by assigning faces to profiles, you can significantly improve the performance and reliability of Kuna's facial recognition feature.