Electronics Weekly
Based on Hailo’s Hailo-8 processor, it is rated at 26Top/s, compared to 13Top/s for the Hailo-8L based ‘AI Kit’ accelerator it announced recently.
“Designed to conform to our HAT+ specification, the AI HAT+ automatically switches to PCIe Gen 3.0 mode to maximise the compute power available in the accelerator,” according to the company. “The 26Top/s model allows you to run multiple networks simultaneously at high frame rates. For instance, you can perform object detection, pose estimation and subject segmentation simultaneously on a live camera feed.”
Unlike the AI Kit, which was a small M.2 board that clipped into Raspberry Pi M.2 accessory HAT, the new accelerator is a full HAT peripheral board (see photo).
Both models use the same software and libraries. “Any neural network model compiled for the Hailo-8L will run smoothly on the Hailo-8; while models specifically built for the Hailo-8 may not work on the Hailo-8L,” said Raspberry Pi, adding: “Alternative versions with lower performance are generally available.”
TensorFlow and PyTorch are amongst supported AI frameworks, and the hardware is integrated into Raspberry Pi’s camera software stack for post-processing tasks such as object detection, image segmentation and pose estimation.
Applications are foreseen in process control, home automation and research.
Find the 26Top/s accelerator here. It will be available for $102.
The organisation also recently revealed a camera with AI hardware built-in.
