AI Processors Consume Less Than 5mW

Kneron’s NPU IP Series artificial intelligence (AI) processors for edge devices includes three devices: the KDP 300 ultra-low power version, the KDP 500 standard version, and the KDP 700 high-performance version. The processors consume under 0.5W, and the KDP 300 designed for facial recognition in smartphones is even less than 5 mW.

 

The series allows ResNet, YOLO and other deep learning networks to run on edge devices under an offline environment. Kneron NPU IP provides complete hardware solutions for edge AI, including hardware IP, compiler, and model compression. It supports various types of Convolutional Neural Networks (CNN) models such as Resnet-18, Resnet-34, Vgg16, GoogleNet, and Lenet, as well as mainstream deep learning frameworks, including Caffe, Keras, and TensorFlow.

Sponsored by Digi-Key

TE Connectivity Horticultural Lighting Solutions Available Now from Digi-Key

TE connector, relay and filter solutions can help create a fine maze network of power-cabling to the lights needed for vertical farming. Their solutions are interchangeable, easy-to-install, and will last for years in variable humidity environments.

 

NPU IP- KDP 300 Ultra-low Power Version

 

KDP 300 supports faster and a more accurate 3D live facial recognition through image analysis from 3D structured light and dual-lens cameras. KDP 300 is also suitable for edge devices that require ultra-low power consumption. The power, including computing and SRAM (Static Random-Access Memory), is less than 5mW.

 

NPU IP- KDP 500 Standard Version

 

KDP 500 can do real-time recognition, analysis, and deep learning for mass faces, hand and body gestures, which is ideal for applications in smart home and smart surveillance. Its computing capacity is up to 152 GOPS (500MHz) (billion operations per second), while sustaining 100mW power consumption.

 

NPU IP- KDP 700 High-performance Version

 

KDP 700 supports more advanced and complex AI computing, as well as deep learning inference for high-end smartphones, robots, drones, and smart surveillance devices. It is currently in the development stage and is expected to offer superior computing capacity with peak throughput up to 4.4 TOPS (1GHz) (trillion operations per second) while keeping the power consumption 300~500mW.

Note 1: Measurement conditions: CNN slice size in 150x150, CNN frame rate at 5fps, and main frequency at 20Mhz.

Note 2: Energy efficiency varies depending on the semiconductor processes. Under 40 nm process, the energy efficiency of Kneron NPU can reach higher than 1.5 TOPS/W.

 

For more information, visit Kneron.

Suggested Articles

PCB layout may seem pretty easy, but doing it well takes a different skill set than design

Eta Compute Inc. has begun shipping its ECM3532 multicore processor for embedded sensor applications.

With coronavirus showing no signs of letting up soon, industry trade shows such as Mobile World Congress are being cancelled.