Secure Entry Systems with id3 Face Recognition on STM32N6

We are proud to announce a new use case developed in partnership with STMicroelectronics: a secure entry system powered by our MicroFace SDK running on the STM32N6 microcontroller with its Neural-ART accelerator.

Ultra-fast on-device face recognition

The solution delivers ultra-fast facial recognition entirely on the edge — no cloud dependency, no servers. The MicroFace SDK detects faces in 8 ms and extracts biometric features in 20 ms, enabling real-time verification while keeping data fully local for maximum privacy and reduced infrastructure costs.

Liveness detection that stops spoofing

What sets this solution apart is the embedded liveness detection using ST's VL53L5CX Time-of-Flight (ToF) sensor with 8x8 multi-zone telemetry. Liveness checks stop fake attempts in just 60 ms, offering far stronger anti-spoofing protection than 2D-only methods — even in poor lighting conditions.

Sensor fusion on a single MCU

The system combines a Sony IMX335 RGB image sensor (5 MP, HDR) with the ToF sensor, all processed locally by the STM32N6. Sensor fusion rapidly detects faces, verifies liveness, and matches biometric features against a local identity database — all on-device.

Optional mobile companion app

The solution can be paired with a mobile app that modernizes user onboarding, providing intuitive face registration and secure transmission of biometric features to the embedded device.

Read the full case study on STMicroelectronics' website.