1768: AI Computer Vision and Motion Analytics Unlocking CCTV Data

Fyma was founded by Karen Burns and Taavi Tammiste. They have created an AI platform that easily plugs into most retail, city, urban, and CCTV cameras, transforming your standard camera feed into deep insights into objects, people, and bahaviours in real-time.

Use cases include The London LegacyDevelopment Corporation, using AI to keep building the Olympic park legacy; major retailers and shopping centres; the Dutch GranPrix; European city governments; urban planners and event spaces.

It's more ethical - the AI is trained to not recognise faces, ever, in the training and deployment stages of AI, so there's no risk of facial/racial profiling. And, Fyma does not store any camera feed data, only the metadata.

For businesses, it's good for unlocking data, getting new insights to create more tailored retail experiences, loyalty schemes, offers, etc., to improve in-store revenue. For venue owners and urban planners, it means getting the insights to adapt and create more accessible, inclusive, and efficient spaces.

For citizens and consumers, it's all about protecting privacy, avoiding data being misused, no racial or facial profiling. Fyma is pushing ahead with AI learning in new ways and focusing on other things (movements, weather, gender, age, etc.).

Taavi Tammiste, CTO and co-founder of Fyma joins me on the Tech Talks Daily Podcast to share the story behind the company.