AI chipmaker Hailo introduces the Hailo-15

With the launch of Hailo-15, the company is setting a new standard in computer vision and deep learning video processing, capable of delivering AI performance in a wide range of applications for different industries.

With Hailo-15, smart city operators can more quickly detect and respond to incidents; manufacturers can increase productivity and machine uptime; retailers can protect supply chains and improve customer satisfaction; and transportation authorities can recognise everything from lost children, to accidents, to misplaced luggage.

“Hailo-15 represents a significant step forward in making AI at the Edge more scalable and affordable,” stated Orr Danon, CEO of Hailo. “With this launch, we are leveraging our leadership in Edge solutions, which are already deployed by hundreds of customers worldwide; the maturity of our AI technology; and our comprehensive software suite, to enable high-performance AI in a camera form-factor.”

The Hailo-15 VPU family includes three variants — the Hailo-15H, Hailo-15M, and Hailo-15L — to meet the varying processing needs and price points of smart camera makers and AI application providers. Ranging from 7 TOPS (Tera Operation per Second) up to an astounding 20 TOPS, this VPU family enables over five times higher performance than currently available solutions in the market, at a comparable price point. All Hailo-15 VPUs support multiple input streams at 4K resolution and combine powerful CPU and DSP subsystems with Hailo’s field-proven AI core.

By introducing superior AI capabilities into the camera, Hailo is addressing the growing demand in the market for enhanced video processing and analytic capabilities at the Edge. With this unparalleled AI capacity, Hailo-15-empowered cameras can carry out significantly more video analytics, running several AI tasks in parallel including faster detection at high resolution to enable the identification of smaller and more distant objects with higher accuracy and fewer false alarms.

As an example, the Hailo-15H is capable of running the state-of-the-art object detection model YoloV5M6 with high input resolution (1280x1280) at real-time sensor rate, or the industry classification model benchmark, ResNet-50, at an extraordinary 700-FPS.

With this family of high-performance AI vision processors, Hailo is also pioneering the use of vision-based transformers in cameras for real-time object detection. The added AI capacity can also be utilised for video enhancement and much better video quality in low-light environments, for video stabilisation, and high dynamic range performance.

Hailo-15-empowered cameras lower the total cost of ownership in massive camera deployments by offloading cloud analytics to save video bandwidth and processing while improving overall privacy due to data anonymisation at the Edge. The result is an ultra-high-quality AI-based video analytics solution that keeps people safer while ensuring their privacy and allows organisations to operate more efficiently, at a lower cost and complexity of network infrastructure.

The Hailo-15 vision processors family, like the Hailo-8TM AI accelerator, which is already widely deployed, are engineered to consume very little power, making them suitable for every type of IP camera and enabling the design of fanless Edge devices. The small power envelope means camera designers can develop lower-cost products by leaving out an active cooling component. Fanless cameras are also better suited for industrial and outdoor applications where dirt or dust can otherwise impact reliability.

“By creating vision processors that offer high performance and low power consumption directly in cameras, Hailo has pushed the limits of AI processing at the Edge,” said KS Park, Head of R&D for Truen. “Truen welcomes the Hailo-15 family of vision processors, embraces their potential, and plans to incorporate the Hailo-15 in the future generation of Truen smart cameras.”

“With Hailo-15, we’re offering a unique, complete and scalable suite of Edge AI solutions,” Danon concluded. “With a single software stack for all our product families, camera designers, application developers, and integrators can now benefit from an easy and cost-effective deployment supporting more AI, more video analytics, higher accuracy, and faster inference time, exactly where they’re needed.”