Monday, July 17, 2023
Imagine a TV that shows tailored recommendations and adjusts the volume for each viewer, or a video doorbell that notifies you when a stranger is at the door. A coffee machine that knows exactly what you want so you only have to confirm. A car that adjusts the seat as soon as you get in, because it knows who you are. All of this and more is possible with Familiar Face Identification, a technology that enables devices to recognize their users and personalize their settings accordingly.
Unfortunately, common methods for Familiar Face Identification are either inaccurate or require running large models in the cloud, resulting in high cost and substantial energy consumption. Moreover, the transmission of facial images from edge devices to the cloud entails inherent risks in terms of security and privacy.
At Plumerai, we are on a mission to make AI tiny. We have recently succeeded in bringing Familiar Face Identification to the edge. This makes it possible to identify users entirely locally — and therefore securely, using very little energy and with very low-cost hardware.
Our solution uses an end-to-end deep learning approach that consists of three neural networks: one for object detection, one for face representation, and one for face matching. We have applied various advanced model design, compression, and training techniques to make these networks fit within the hardware constraints of small CPUs, while retaining excellent accuracy.
In his talk at the tinyML EMEA Innovation Forum, Tim de Bruin presented the techniques we used to make our Familiar Face Identification solution small and accurate. By enabling face identification to run entirely on the edge, our solution opens up new possibilities for user-friendly and privacy-preserving applications on tiny devices.
These techniques are explained and demonstrated using our live web demo, which you can try for yourself right in the browser.