RKNN single thread for Orange Pi 3B
The code is copied somewhere and modified to use on Orange Pi 3B since this board has only single thread.
Depend on your Python version, you need to download the RKNN Toolkit Lite 2 from this https://github.com/rockchip-linux/rknn-toolkit2 Included in this github is the wheel version 1.5.0 for Python 3.9. Install it by typing command
pip install rknn_toolkit_lite2-1.5.0-cp39-cp39-linux_aarch64.whl
The model yolov5s.rknn is converted from yolov5s.onnx using rknn_toolkit_lite2-1.5.0, maybe 1.5.2 works.
Maybe you also need to copy the file librknnrt.so to /usr/lib too, and install opencv module
pip install opencv-python
Check your Camera used in /dev/video0 or /dev/video1, and edit the code in main.py to the corresponance camera
cap = cv2.VideoCapture(0)
then run testing with images and dataset included
python deploy.py
if everthing run well, run the camera
python cameera.py
The rkcat.sh can be running with
sudo bash rkcat.sh
to check the NPU performance
Here is demo i run on Orange Pi 3B with USB Camera, it can reach 20fps
