Are there any measures in place to ensure demographic estimates are not biased or discriminatory?
Aura Vision starts with a trained generic AI model that is fine-tuned per store to improve accuracy and reduce bias.
All data is fully anonymized and aggregated to protect privacy.
Models are regularly reviewed and re-trained if accuracy drops.
The system follows strict GDPR and ethical standards to prevent discrimination.
Related Articles
How do you train your AI?
During initial setup, Aura Vision’s team of Computer Vision experts will audit and fine-tune the AI models to improve the accuracy of people counts, age and gender demographics and train the algorithms to segment customers and staff based on uniform ...
How is the accuracy of entry counts evaluated?
We conduct initial audits for each location to ensure the accuracy of counts meets our strict requirements. In the rare instance where a location fails an audit, it is usually because the entrance has been obstructed — possibly by a product display ...
Can demographics be switched on for individual stores / groups of stores?
Demographics are available at organisation-level in general but If different demographics requirements are needed we recommend separating stores into different dashboard organisations. Just drop us an email at support@auravision.ai and we'll work ...
How accurate is Age recognition?
Age recognition is typically between 85% to 90% We segment age data and into the following seven categories: under 16 16 to 24 25 to 34 35 to 44 45 to 54 55 to 64 65+
How does it recognise age and gender?
We use an AI powered algorithm that has been trained on massive datasets. It looks at multiple characteristics including body shape and clothing, and has passed extensive accuracy tests.