Kinara, Inc., a developer of unrivaled Edge AI solutions that accelerate and optimise real-time decision making, today announced that it has partnered with Arcturus Networks Inc., a provider of edge AI analytics and enablement for smart city applications, to help customers deliver high-performance, cost-optimised AI solutions.
The partnership combines the Kinara Ara-1 Edge AI processor with Arcturus Brinq edge AI and vision analytics software to drive detection, tracking and characterisation solutions. The partnership delivers the critical software and hardware platforms required by OEMs to build sophisticated real-time edge applications for public safety, transportation, healthcare, retail, and industrial markets. For example, Kinara and Arcturus can deliver a real-time solution for road condition monitoring, allowing public transportation buses retrofitted with smart cameras to detect road obstructions such as potholes and report them directly to city operations.
The Ara-1 Edge AI processor delivers a ground-breaking combination of performance, power, and price for integrated cameras and edge servers. The Kinara architecture can effectively run multiple models with low latency and zero switching overhead, generating results faster and more accurately. Kinara complements its processing technology with a comprehensive set of development tools with built-in quantiser, simulator, profiler, and compiler to enable its customers to optimise and easily run their neural network models using the Ara-1 Edge AI processor.
Brinq is an edge AI software suite that combines vision algorithms, state-of-the art inference, analytics applications, and tools to help customers fine-tune ML models. The software includes industry leading Multiple Object Tracking (MOT) technology that recently garnered Arcturus among top honors at the global MOT Challenge benchmark competition. Supported hardware includes the Arcturus Atlas scalable edge inference platform using NXP Semiconductors’ i.MX 8M Plus SoC and further accelerated using Kinara’s Ara-1. The combination of Brinq software, Atlas hardware and Kinara’s Ara-1 delivers a complete high-performance, turn-key, AI platform that de-risks development, reduces time-to-market, and lowers total cost of ownership for OEMs.
“Our Brinq software provides a flexible edge AI pipeline architecture that makes it easy to integrate a variety of compute units such as CPUs, GPUs, or NPUs,” says David Steele, director of innovation at Arcturus Networks. “By using Kinara’s Ara-1 processor we can deliver more sophisticated smart city analytics that achieve higher throughput and improved accuracy. This allows us to enable a new generation of edge AI applications while achieving excellent price/performance value for our customers.”
“Arcturus uses its Atlas platform with NXP’s i.MX 8M Plus applications processor to provide excellent in-device inference performance for standard classification and detection algorithms, and we can cost-effectively add our Ara-1 edge AI accelerator to the Atlas platform to deliver the ability to scale up to an additional 20x higher inference performance for more complex models and applications demanding higher resolution and higher accuracy video processing,” says Markus Levy, VP business development of Kinara.
“It’s great to see our partners, Arcturus and Kinara, working together to add significant value to the NXP ecosystem,” says Ali Osman Ors, director, AI ML strategy and technologies at NXP Semiconductors. “Their collaboration will help to extend the benefits of products like the i.MX 8M Plus and provide our mutual customers further scalability in developing real-world AI solutions.”
Download the technical paper “Practical Requirements for Mission-Critical Edge AI Applications” to get additional insight into real-world AI application development.
Follow us and Comment on Twitter @TheEE_io