Through the partnership, the companies have said that they want to see an order of magnitude increase in performance and power efficiency for deep learning in edge devices compared to current GPU-based implementations.
While there is a real and sustained push towards widespread adoption of Artificial Intelligence (AI) in consumer devices, cloud-based deep learning on battery-powered devices is confronted with issues, including latency, security and the need for a constant, reliable internet connection.
Implementing intelligence on the device itself – or at the edge – is intended to eliminate all of these issues.
Highly efficient computer vision processors will be necessary to meet the stringent power requirements, while specialised deep learning software will be crucial in delivering the accuracy and performance needed for cloud-based systems.
Targeting embedded devices, Brodmann17 has developed a specialised deep learning technology for visual recognition aimed at edge-based artificial intelligence.
Using patent-pending techniques, the company’s deep learning architecture generates smaller neural-networks that are faster and more accurate than any other network generated on the market. Through collaboration with Brodmann17, licensees of the CEVA-XM platforms and their customers will be able to use Brodmann17’s deep learning object detection that achieves state of the art accuracy on the CEVA-XM at a rate of 100 frames per second which is significantly faster than existing GPU-based solutions.
“Our patent-pending deep learning vision software is a perfect fit for the many CEVA customers and OEMs using CEVA-XM platforms to add intelligence to their devices,” said Adi Pinhas, CEO of Brodmann17. “This first-of-its-kind combination of hardware and software achieves real-time performance that supports multi-cameras with a single DSP or higher resolutions.”
IlanYona, vice president of the vision business unit at CEVA added, “Brodmann17’s deep learning software provides the capability to create extremely light, accurate and flexible networks, trained from the ground up with embedded in mind.”
