Hailo live demonstrations at embedded world 2024
Hailo is staging live demonstrations of Hailo-8 accelerator and Hailo-15 vision processor integrations at the embedded world 2024 Exhibition and Conference (April 9th - April 11th) at Stand 126 in Hall 1 of Messe Nuremberg in Nuremberg, Germany.
Hailo is hosting live demos of Industrial Automation and Security applications developed by leading providers and empowered by the Hailo-8 AI accelerator as well as the Hailo-15 AI vision processor that has been nominated for the 2024 Embedded Awards in the Embedded Vision category.
According to Hailo CEO Orr Danon, Hailo-15 is specifically designed to empower smart cameras with unprecedented AI capacity. Show attendees can see Hailo-15 in action, running analytics tasks in real-time and high resolution, while managing some advanced AI-powered image enhancement and low-light denoising capabilities. Additionally, Hailo will be demonstrating multiple customer and partner demos at the show, including Accenture, B&R, Deep Vision Consulting, EBV, iRider, Keba, Lex Systems, LinkerVision, MVTec, Raspberry Pi, Schneider Electric, SolidRun, System Electronics, Truen, and Wings for Aid, all displaying their own Hailo-empowered products.
“embedded world 2024 is a key opportunity for us to present the capabilities of our portfolio of edge AI solutions, in a wide range of embedded applications in cooperation with leading partners who have incorporated our processors into their product solutions,” Danon said. “Bolstered by the success of our products and the support of our investors, Hailo is ready to bring to market a steady stream of diverse solutions, from ARAS (iRider bike), to cameras of all types (ITS, security, industrial), and everything in between.”
The Hailo-8 AI accelerator is the world’s most efficient AI processor in terms of both performance to cost and performance-to-power consumption ratios. The compute power enabled by the Hailo-8 accelerators ranges from 13 TOPS for the Hailo-8L, designed for entry-level AI applications, all the way to 200 TOPS for the high-performance Century PCIe card, which enables a large number of streams on large-scale video management systems.
Delivering unprecedented AI performance in a camera power envelope, the Hailo-15 System-on-a-Chip (SoC) combines Hailo’s patented and field-proven AI inferencing capabilities with advanced computer vision engines, generating premium image quality and advanced video analytics. The unprecedented AI capacity can be used for both AI-powered image enhancement such as low-light denoising, image stabilisation, HDR, auto focus and more, and processing of multiple complex deep learning AI applications at full scale and at superior efficiency.
All the Hailo processors, including the newly launched Hailo-10 use the same neural core architecture which takes advantage of the core properties of neural networks and allows edge devices to run deep learning applications at full scale more efficiently, effectively, and sustainably than other AI chips and solutions, while significantly lowering costs. They also share the same robust software stack which allows developers to easily integrate their AI applications and deploy them on a wide range of edge platforms from cars and off-road vehicles, through smart cameras and robots to personal computers.
Last week, Hailo announced the extension of its series C fundraising round with the successful closing of an additional $120 million that brings the company’s total funding to more than $340 million. At the same time, the company announced the introduction of its innovative Hailo-10 high-performance generative AI (GenAI) accelerator that ushers in an era where users can own and run GenAI applications locally without registering to cloud-based GenAI services.
Enabling GenAI at the edge ensures continuous access to GenAI services, regardless of network connectivity; obviates network latency concerns, which can otherwise impact GenAI performance; promotes privacy by keeping personal information anonymised; and enhances sustainability by reducing reliance on the substantial processing power of cloud data centres.