Hailo claims AI-centric vision processor is a first for intelligent cameras
For cameras used in intelligent, connected areas, such as smart cities, factories and retail locations, the Hailo-15 vision processor units (VPUs) are believed to be the first designed for integration directly into intelligent cameras. According to Hailo, they deliver unprecedented video processing and analytics at the edge and redefine the smart cameras with a new standard in computer vision and deep learning video processing.
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 identification of smaller and more distant objects with higher accuracy and less false alarms.
As an example, the Hailo-15H is capable of running the object detection model YoloV5M6 with high input resolution (1280 x 1280) at real time sensor rate, or the industry classification model benchmark, ResNet-50, at 700 frames per second.
This will enable, for example, smart city authorities can more quickly detect and respond to incidents. Another example is that manufacturers can increase productivity and machine uptime, retailers can protect supply chains and improve customer satisfaction and transportation authorities can recognise incidents such as lost children, traffic accidents and misplaced luggage.
The Hailo-15 VPU family includes the Hailo-15H, Hailo-15M, and Hailo-15L which have varying processing levels and price points, ranging from 7Toperations per second to 20T operations per second. The VPU family enables over five times higher performance than currently available solutions in the market, at a comparable price point, claimed Hailo. All three Hailo-15 VPUs support multiple input streams at 4K resolution and combine a powerful CPU and DSP subsystems with Hailo’s AI core.
As well as vision-based transformers in cameras for real time object detection, the added AI capacity can be used for video enhancement and to improve 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, explained Hailo. The result is an ultra-high-quality AI-based video analytics solution that keeps people safer, while ensuring their privacy and allows organizations to operate more efficiently, at a lower cost and complexity of network infrastructure.
“Hailo-15 represents a significant step forward in making AI at the edge more scalable and affordable,” stated Orr Danon, Hailo’s CEO.
“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,” he added.