Maxim Allies with Xailient to Provide IoT Face Detection
Maxim Integrated and Xailient have reported that Maxim Integrated’s MAX78000 ultra-low-power neural-network microcontroller detects and localizes faces in video and images using Xailient’s proprietary Detectum neural network.
Xailient’s neural network draws 250x lower power (at just 280 microJoules) than conventional embedded solutions, and at 12 milliseconds (ms) per inference, the network performs in real-time and is faster than the most efficient face-detection solution available for the edge.
“With the Xailient Detectum neural network, the MAX78000 is capable of both classification and localization, so in addition to seeing faces in the image or video you can also determine where those faces are in the image’s field of view,” said Robert Muchsel, Maxim Integrated Fellow and architect of the MAX78000 microcontroller. “Advanced applications include person, vehicle and object counting, presence or obstruction detection, as well as path mapping and footfall heatmaps.”
“AI is on track to be the second-largest carbon-emitting industry,” said Shivy Yohanandan, Xailient CTO and inventor of Xailient’s Detectum neural network technology. “Replacing 14 legacy Internet protocol cameras that use traditional cloud AI with edge-based cameras equipped with the Maxim Integrated MAX78000 paired with Xailient’s neural network has the equivalent carbon impact of taking one gasoline-powered car off the road.”
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.
In addition to supporting standalone applications, Maxim Integrated’s microcontroller paired with Xailient’s neural network improves overall power efficiency and battery life of hybrid edge/cloud applications that employ a low-power ‘listening’ mode which then awakens more complex systems when a face is detected.
Xailient’s Detectum neural network includes focus, zoom and visual wake-word technologies to detect and localize faces in video and images at 76x faster rates than conventional software solutions, at similar or better accuracy.
In addition, the flexible network can be extended to applications other than facial recognition, such as livestock inventory and monitoring, parking spot occupancy, inventory levels and more.
