Adoption of facial recognition, powered by artificial intelligence (AI), is growing for a couple of reasons: faster response and higher accuracy. Maxim Integrated Products, Inc. and Xailient Inc. have developed an IoT face detection solution that detects and localizes faces in video and images at 12 milliseconds (ms) per inference. The solution combines Maxim Integrated’s MAX78000 ultra-low-power neural-network microcontroller (MCU) with Xailient’s proprietary Detectum neural network, which optimizes the computational efficiency and low-power sleep modes provided by the MAX78000.
“Speed is a significant factor for AI because with faster inferencing, you can react in real time or quickly average multiple inferences to improve accuracy,” said Maxim Integrated.
(Source: Maxim Integrated Products Inc.)
Battery-powered AI systems that require face detection, such as home cameras, industrial grade smart security cameras and retail solutions, require a low-power solution to provide the longest possible operation between charges, said Maxim. Xailient’s neural network is said to draws 250× lower power (at 280 microJoules) than conventional embedded solutions.
Together, the devices improve overall power efficiency and battery life of hybrid edge/cloud applications that use a low-power ‘listening’ mode, which awakens more complex systems when a face is detected. The solution can extend the operating time of these coin cell battery-powered applications for many years, according to the companies.
Xailient’s Detectum neural network includes focus, zoom, and visual wake-word technologies to detect and localize faces in video and images at 76× faster rates than conventional software solutions, at similar or better accuracy, according to the companies. In addition to facial recognition, the neural network can be used in a variety of other applications such as livestock inventory and monitoring, parking spot occupancy, and inventory levels.
It is also better for the environment. Xailient said the solution can reduce carbon emissions by replacing 14 legacy internet protocol cameras that use traditional cloud AI with edge-based cameras using the MAX78000 paired with its neural network. The equivalent carbon impact is taking one gasoline powered car off the road, according to Dr. Shivy Yohanandan, Xailient CTO and inventor of Xailient’s Detectum neural network technology.
The MAX78000 is priced at $8.50 (1000-up, FOB USA) at Maxim’s website and is also available from authorized distributors. The MAX78000EVKIT# evaluation kit is priced at $168. The Detectum neural network, series models, tools, and services as well as focus, zoom and visual wake word technologies are available directly from Xailient
about Maxim IntegratedMaxim Integrated Products