This power consumption is 100 times lower than that of conventional software solutions, making it the most cost-effective and efficient IoT people detection solution available. The low-power network provides extended operation for battery-powered IoT systems that require human presence detection, such as building energy management and smart security cameras.
The low-power, neural network-accelerated MAX78000 microcontroller performs AI inference at less than one-hundredth the energy of conventional software solutions, improving the runtime of battery-powered edge AI applications. The mixed-precision VWW network is part of the Aizip Intelligent Vision Deep Neural Network (AIV DNN) series for image and video applications and was developed using Aizip's proprietary design automation tools to achieve over 85% accuracy for human presence.
Key benefits
- Longer battery life: The efficient AI model and low-power system-on-a-chip (SoC) reduce inference energy to 0.7 mJ, enabling 13 million inferences on a single AA/LR6 battery.
- Cost-effective edge intelligence: Extreme model compression enables precise intelligent vision with a low-cost AI-accelerated microcontroller and low-cost image sensors.
