Raspberry Pi released AI kit, which gives unlimited imagination about how we can use Raspberry Pi now! It's time to elevate our IoT projects to next level!
This wiki demonstrates an object detection model using YOLOv8 on reComputer R1000 with Raspberry-pi-AI-kit Acceleration. The Raspberry Pi AI Kit enhances the performance of the Raspberry Pi and unlocks its potential in artificial intelligence and machine learning applications, like smart retail, smart traffic, and more. Although the Raspberry AI Kit is designed for Raspberry Pi 5, we have experimented with it on our CM4-powered edge gateway. Excited about turning our edge device into an intelligent IoT gateway!
Step 1: Install AI kit Update system
Open the terminal on the reCompuer R1000, and input command as follows to update your system.
sudo apt update
sudo apt full-upgrade
Set PCIe to gen3 Open terminal on the reCompuer R1000, and input command as follows to config reCompuer R1000.
sudo raspi-config
Select option "6 Advanced Options"
Then select option "A8 PCIe Speed"
Choose "Yes" to enable PCIe Gen 3 mode
Click "Finish" to exit
Open terminal on the reCompuer R1000, and input command as follows to install Hailo software.
sudo apt install hailo-all
sudo reboot
Check Software and Hardware Open terminal on the reCompuer R1000, and input command as follows to check if hailo-all have been installed.
hailortcli fw-control identify
The right result show as bellow:
Open terminal on the reCompuer R1000, and input command as follows to check if hailo-8L have been connected.
lspci | grep Hailo
The right result show as bellow:
Open terminal on the reCompuer R1000, and input command as follows to run YOLOv8.
git clone https://github.com/Seeed-Projects/Running-YOLOv8-Object-Detection-on-reComputer-R1000-CM4-Powered-Edge-Gateway-with-Hailo-8L.git
cd Running-YOLOv8-Object-Detection-on-reComputer-R1000-CM4-Powered-Edge-Gateway-with-Hailo-8L/object_detection_benchmark/Yolov8-with-AIkit
bash ./run.sh
Result We compared the inference speed of YOLOv8 for object detection before and after acceleration using the AI kit. The results show that before acceleration, the inference speed was only 0.75 FPS, whereas after acceleration, it reached 29.5 FPS.
Project Outlook
Comments