Edge Impulse brings NVIDIA-powered models to edge MCUs, MPUs, and AI accelerators
Edge Impulse, a platform designed for developing, refining, and deploying machine learning models and algorithms to edge devices, has introduced a new set of tools created using NVIDIA’s Omniverse and AI platforms. This initiative aims to extend the reach of AI models to a previously untapped category of devices situated at the edge.
Presently, Edge Impulse offers a solution that streamlines and speeds up the utilization of extensive NVIDIA GPU-trained models on MCUs (Microcontroller Units) and MPUs (Microprocessor Units) equipped with AI accelerators.
According to the company, users now have access to a collection of NVIDIA production-tested pretrained models directly within Edge Impulse’s platform. Additionally, Edge Impulse’s EON Tuner aims to simplify the process of selecting the most suitable model for each specific application.
Through the collaboration between Edge Impulse and the NVIDIA TAO toolkit, engineers can develop tailored, ready-for-production computer vision models, the company notes. These models can deploy to hardware optimized for edge computing, such as the Arm Cortex-M based NXP I.MXRT1170, Alif E3, STMicro STM32H747AI, and Renesas CK-RA8D1.
Additionally, the Edge Impulse platform lets users incorporate their own custom data with GPU-trained NVIDIA TAO models like YOLO and RetinaNet, optimizing them for deployment on efficient, cost-effective edge devices, including MCUs, MPUs, and accelerators.
This advancement also facilitates the deployment of extensive NVIDIA models to Arm-based devices, expanding the range of hardware capable of leveraging top-tier AI and ML models.
“The advent of generative AI and the growth of IoT deployments means the industry must evolve to run AI models at the edge,” says Paul Williamson, senior vice president and general manager, IoT line of business at Arm.
“NVIDIA and Edge Impulse have now made it possible to deploy state-of-the-art computer vision models on a broad range of technology based on Arm Cortex-M and Cortex-A CPUs and Arm Ethos-U NPUs, unlocking a multitude of new AI use cases at the edge.”
Edge Impulse has developed applications for synthetic data and testing environments for the edge with NVIDIA Omniverse.
“Working closely with NVIDIA has enabled us to significantly expand the practical applications of AI on the edge for critical business use cases in industrial productivity, healthcare, and much more. For the first time, NVIDIA’s state-of-the-art machine learning research and model architectures can be deployed on any device under the sun, from the smallest microcontrollers to the latest GPUs and neural accelerators,” adds Jan Jongboom, co-founder and CTO of Edge Impulse.
“NVIDIA Omniverse and TAO have incredibly simplified the creation of all computer vision models, including the latest generative AI models,” said Deepu Talla, vice president of robotics and edge computing at NVIDIA. “Edge Impulse is integrating this powerful capability into easy-to-use workflows for the hundreds of billions of IoT and edge devices, including MCUs, accelerators and CPUs.”
Read more:
Edge Impulse incorporates Nvidia TAO Toolkit for edge AI; partners with AWS
Edge Impulse, Particle.io unite to support Photon 2 platform and improve IoT development
Article Topics
edge computing | edge devices | Edge Impulse | MCU | Nvidia
Comments