Himax and Edge Impulse Launch the First Ultralow Power AI Vision and Sensor Fusion Solution
Himax Technologies, Inc., a leading supplier and fabless manufacturer of display drivers and other semiconductor products, announced a collaboration with Edge Impulse that enables the rapid development and deployment of machine learning (ML) models on the Himax HX6537-A WE-I Plus AI processor and Google TensorFlow Lite for Microcontrollers framework, allowing low-power, memory-constrained, and remote edge devices to detect complex motion, recognize sounds and classify images at sub-mW power consumption. The AI vision and sensor fusion solution is ideal for predictive maintenance, condition monitoring, asset tracking, and occupancy detection applications where power efficiency, bandwidth optimization and privacy are key requirements.
“Himax and Edge Impulse are teaming up to help developers create amazing new user experiences with secure, private, easy to use embedded development tools,” said Jordan Wu, President and Chief Executive Officer at Himax. “Offered at no charge to individual Himax developers, Edge Impulse helps simplify the creation of neural networks across a range of Himax products, including AI processors and always-on sensors, all operating at ultralow power. Edge Impulse makes it easy for embedded developers to develop TinyML models on real sensor data – from collecting data in the field to designing signal processing and ML pipelines, and to deploying models efficiently on Himax HX6537-A WE-I Plus processor with TensorFlow Lite for Microcontrollers.”
“Himax WE-I Plus and Edge Impulse end-to-end platform enable an easy TinyML development with TensorFlow on an ultralow power device,” said Pete Warden, Technical Lead of TensorFlow Lite for Microcontrollers at Google.
By using Edge Impulse and Himax, engineers can generate and export the models directly to the devices with a single button press, and in seconds, not days. These resulting inferences fully utilize the latest neural networks and Edge Impulse EON™ technology to ensure that they run as fast and energy-efficient as possible.
“The industrial, enterprise, and consumer applications of embedded ML are endless, enabling highly customizable, data-driven value for customers. Edge Impulse support for Himax AI processors and always-on vision sensors is a game changer for these markets, unlocking new value for customers by enabling ML to be embedded anywhere at extremely low-power,” said Zach Shelby, co-founder, and CEO at Edge Impulse.
Recommended AI News: Emirates NBD Boosts Digital User Engagement with Lemnisk