Hailo Introduces Hailo-15: AI-Centric Vision Processors for Next-Generation Intelligent Cameras
January 30, 2024
Hailo announced its new Hailo-15 family of high-performance vision processors, designed for direct integration into intelligent cameras for video processing and analytics at the edge.
The Hailo-15 VPU family includes three variants — the Hailo-15H, Hailo-15M, and Hailo-15L — providing varying processing needs and price points for smart camera makers and AI application providers. Per the company, the VPU family ranges from 7 TOPS (Tera Operation per Second) up to 20 TOPS, enabling over 5x higher performance than currently available solutions in the market. All Hailo-15 VPUs support multiple input streams at 4K resolution and combine CPU and DSP subsystems with Hailo’s AI core.
By supporting AI capabilities into cameras, Hailo is addressing the growing demand in the market for enhanced video processing and analytic capabilities at the edge. According to the company, the AI capacity helps Hailo-15-empowered cameras to carry out more video analytics, running several AI tasks in parallel including fast detection at high resolution to enable the identification of small, distant objects with high accuracy and less false alarms.
For example, the Hailo-15H is capable of running the state-of-the-art object detection model YoloV5M6 with high input resolution (1280×1280) at real time sensor rate, or the industry classification model benchmark, ResNet-50, at an extraordinary 700 FPS.
Hailo is also pursuing the use of vision-based transformers in cameras for real-time object detection. The added AI capacity can also be utilized for video enhancement and improved video quality in low-light environments, for video stabilization, and high dynamic range performance.
Hailo-15-empowered cameras lower the total cost of ownership in camera deployments by offloading cloud analytics to save video bandwidth and processing, while improving overall privacy due to data anonymization at the edge.
The Hailo-15 vision processors family, like the Hailo-8 AI accelerator, are engineered to consume very little power, making them suitable for IP cameras 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 suitable for industrial and outdoor applications where dirt or dust can otherwise impact reliability.
For more information, visit hailo.ai.