At Embedded World 2025 in Nürnberg, Hailo presented its latest advancements in AI processing, focusing on the Hailo-15 AI Vision Processors. citeturn0search0 These processors are designed to enhance edge devices with robust AI capabilities, enabling real-time data processing directly on the device.
—
Synaptics is my Embedded World 2025 video coverage sponsor, check out my Synaptics videos here: https://www.youtube.com/playlist?list=PL7xXqJFxvYvhAbQoe9YN4c84SqXxIY3fQ
—
One notable demonstration featured the integration of the Hailo-15 processor into drones developed in collaboration with System Electronics. These drones showcased advanced object and face detection capabilities, emphasizing the processor’s high performance and low power consumption—critical factors for extending battery life in aerial applications.
Another highlight was the NG Cam smart camera, equipped with the Hailo-15 processor. This device demonstrated the ability to perform zero-shot learning using the CLIP model, allowing it to recognize and describe objects without prior specific training. For instance, the camera could identify a person wearing glasses by processing the input text “man with glasses,” showcasing the versatility of AI applications in real-world scenarios.
In partnership with SolidRun, Hailo introduced a System on Module (SoM) featuring the Hailo-15H processor. This compact module delivers up to 20 TOPS (Tera Operations Per Second) and supports high-resolution imaging, including 4K video at 60 frames per second. Its robust design ensures reliable operation across various industrial environments, making it suitable for applications in security, industrial automation, and smart cities.
The Hailo-15 processor also enhances image quality through AI-driven features such as low-light denoising and high dynamic range (HDR) processing. Demonstrations highlighted the processor’s ability to improve image clarity in low-light conditions and manage high-contrast scenes effectively, ensuring that subsequent analytics, like optical character recognition (OCR), yield more accurate results.
Hailo’s commitment to edge AI extends to the automotive sector. At CES 2025, they showcased integrations of their processors in advanced driver-assistance systems (ADAS), utilizing multiple sensor modalities, including cameras, LiDAR, and radar. These integrations aim to enhance vehicle safety and autonomy by enabling real-time perception and decision-making on the edge.
Furthermore, Hailo’s processors are facilitating the deployment of large language models (LLMs) on edge devices. The upcoming Hailo-10 accelerator, expected to be available later this year, is designed to run models like GPT and Stable Diffusion directly on devices, ensuring low-latency performance and enhanced privacy by keeping data processing local.
Through collaborations with various partners, Hailo continues to push the boundaries of AI at the edge, offering scalable and efficient solutions across multiple industries. Their participation in events like Embedded World 2025 underscores their dedication to advancing edge AI technologies.
Check out all my Embedded World 2025 videos in this playlist: https://www.youtube.com/playlist?list=PL7xXqJFxvYvjgUpdNMBkGzEWU6YVxR8Ga
This video was filmed using the DJI Pocket 3 ($669 at https://amzn.to/4aMpKIC using the dual wireless DJI Mic 2 microphones with the DJI lapel microphone https://amzn.to/3XIj3l8 ), watch all my DJI Pocket 3 videos here https://www.youtube.com/playlist?list=PL7xXqJFxvYvhDlWIAxm_pR9dp7ArSkhKK
Join https://www.youtube.com/charbax/join for Early Access to my videos and to support my work, or you can Click the “Super Thanks” button below the video to send a highlighted comment!
