Convert Yolov8 to CoreML and use it on iPhone

Use object detection AI on your smartphone

Yolov8 is a popular object detection model.
In this article, we will make Yolov8 available for iOS.

GitHub - ultralytics/ultralytics: NEW - YOLOv8 🚀 in PyTorch > ONNX > CoreML > TFLite

Method

Install Yolov8 and download model checkpoint.
(A list of model checkpoints can be found in the ultralytics repository above.)

pip install ultralytics
wget https://github.com/ultralytics/assets/releases/download/v0.0.0/yolov8s.pt

If you get the following error, change the following line in yolo/engine/exporter.py of the ultralytics library (line 390).

# f = self.file.with_suffix(‘.mlmodel’)
# ↓↓↓
f = self.file.with_suffix(‘.mlpackage’)

A .mlpackage file (CoreML) is saved in the same directory as the pt file.

The sample to run with the camera of iOS is as follows.

GitHub - john-rocky/Yolov8-RealTime-iOS

🐣

I’m a freelance engineer.
Work consultation
Please feel free to contact us with a brief development description.
rockyshikoku@gmail.com

I am creating applications using machine learning and AR technology.

I send machine learning / AR related information.

GitHub

Twitter
Medium