FURK4NGG
28 lines
1.1 KiB
Markdown
28 lines
1.1 KiB
Markdown
## 👀 run-cv-model-in-rp5 Overview 3/3
|
|
<h1 align="center">Run computer vision model in Rasberry Pi 5 AI HAT</h1>
|
|
|
|
|
|
## 🔎 Preparation
|
|
Open your rasberry pi and follow these steps:
|
|
1. `sudo raspi-config`
|
|
2. `Advanced Options → PCIe Speed → Yes / Enable`
|
|
5. `Interface Options → I2C → Yes / Enable`
|
|
**For connect the rasberry pi remotely:**
|
|
1. `Interface Options → SSH → Yes / Enable`
|
|
2. `Interface Options → RPI Connect|VNC → Yes / Enable`
|
|
|
|
|
|
`/home/$USER/Desktop/yolo_hailo/yolov11n.hef`
|
|
`Write each object class[rock,paper,scissors] on a new line in your that file:'/home/$USER/Desktop/yolo_hailo/labels.txt'`
|
|
|
|
## 📦 Setup
|
|
1. `cd /home/$USER/Desktop/yolo_hailo`
|
|
2. `sudo apt update`
|
|
3. `sudo apt install -y python3-pip python3-libcamera python3-kms++ python3-prctl libatlas-base-dev`
|
|
4. `git clone -b next https://github.com/raspberrypi/picamera2.git`
|
|
5. `cd picamera2`
|
|
6. `pip install -e . --break-system-packages`
|
|
7. `cd picamera2/examples/hailo`
|
|
8. `python detect.py --model /home/$USER/Desktop/yolo_hailo/yolov11n.hef --labels /home/$USER/Desktop/yolo_hailo/labels.txt`
|
|
|