Dark Mode Light Mode

Keep Up to Date with the Most Important News

By pressing the Subscribe button, you confirm that you have read and are agreeing to our Privacy Policy and Terms of Use
Follow Us
Follow Us
Login Login

Hailo and Raspberry Pi work together on the AI extension kit.

Also, an AI computer is the newest iteration of the Raspberry Pi, the little but powerful computer that has become very famous among tech enthusiasts and industrial firms. For the Raspberry Pi 5, the firm unveiled the AI Kit, a $70 expansion kit including a neural network inference accelerator for local inferencing.

This new extension module uses the Raspberry Pi’s HAT+ extension card. Hardware Attached on Top, or HAT, is a clever acronym the business has been using for extra cards you attach to a standard Raspberry Pi.

The HAT+ extension card includes the M.2 slot, a common extension slot for PC components. For readers who are interested in the technical details: Operating at 8 GB/s, a single-lane PCIe 3.0 interface links this slot to the Raspberry Pi.

Advertisement

The business intends to fight Nvidia; hence, it has teamed up for the kit with Hailo, an AI chip startup that just received $120 million. Hailo specializes in processors designed to execute artificial intelligence tasks on edge devices, from autos to smart cameras to robots to Raspberry Pi devices.

The Raspberry Pi is using the Hailo-8L accelerator module for the AI Kit. The HAT+ can easily connect to this entry-level module in an M.2 form factor.

After setting everything up, you receive a Raspberry Pi 5 capable of 13 tera-operations per second (TOPS). While it may not be as powerful as an Nvidia GPU, its cost is affordable and it utilizes the Raspberry Pi’s original 27W power supply.

In terms of software, the most recent Raspberry Pi OS automatically recognises the Hailo module, enabling the OS and applications running on it to use it instantly.

Raspberry Pi has also revised its range of camera programs to enable neural network inferencing inside the camera pipeline. You can use it for object recognition (“this is a car”), semantic segmentation (“these three things are moving vehicles”), instance segmentation (“these three moving vehicles are a truck, a red car, and a blue car”), posture estimation, and face landmarking.

Those are just illustrations of what you could achieve with a Raspberry Pi fitted with the AI kit and a first-party or third-party camera. Still, the Hailo chip finds use outside of cameras.

Seeing the Raspberry Pi community develop fresh applications for this package will be fascinating. Currently, consumers of the Raspberry Pi have to decide what they want to achieve with this AI extension kit; it is just a tool.

Keep Up to Date with the Most Important News

By pressing the Subscribe button, you confirm that you have read and are agreeing to our Privacy Policy and Terms of Use
Add a comment Add a comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Previous Post

Fivetran starts offering a service for managing data lakes.

Next Post

Dive's computational fluid dynamics modelling service moves to the cloud.

Advertisement