A Paris-based AI startup has developed a gesture recognition algorithm to predict potential shoplifting incidents in retail stores. Veesion’s technology analyzes footage from existing security cameras to detect suspicious movements, flagging potential theft for store security personnel.
How Veesion’s Technology Works
Veesion’s algorithm focuses on body movements rather than physical characteristics, addressing concerns about bias and privacy. The system identifies humans in the camera feed, tracks their movements, and recognizes objects such as merchandise, carts, and bags. When suspicious behavior is detected, a video clip is sent to security staff for review.
Impact on Retailers
The startup claims its technology has been deployed in 5,000 stores across Europe, Canada, and the US. One US client reported cutting losses from the health and beauty section in half within three months of implementation. Veesion’s CEO, Benoît Koenig, stated that over 85% of alerts generated by the system are marked as relevant by store operators.
Benefits for Retailers
Veesion’s technology aims to create a more pleasant shopping experience by reducing the need for extreme security measures like locking up items. The system can also help deter employee theft, with Koenig noting that the presence of AI monitoring can have an “internal deterrent effect.”
Future Developments
The company is exploring additional retail applications, including detecting improper scans at self-checkout and slip-and-fall incidents. With a recent $43 million Series B funding round, Veesion is expanding its presence in the US market.
As retailers continue to grapple with shoplifting, Veesion’s technology offers a potential solution that balances security with customer experience. By leveraging existing security infrastructure and focusing on behavior rather than appearance, the startup aims to reduce retail losses while maintaining a welcoming store environment.
