search

ultralytics / yolo-ios-app

Ultralytics YOLO iOS App source code for running YOLOv8 in your own iOS apps 🌟
https://ultralytics.com/yolo
GNU Affero General Public License v3.0
162 stars 29 forks source link

yolov8-seg ios app #11

Open Greywan opened 7 months ago

Greywan commented 7 months ago

Thanks for your efforts, is there any plans to launch an ios app project for yolov8-seg? Or modification of the current project?

glenn-jocher commented 7 months ago

@Greywan hey there! 🙌 Thanks for your interest and kind words. As of now, we don't have immediate plans for launching or modifying an iOS app specifically for yolov8-seg. However, the community's interest drives many of our decisions, and we're always open to exploring new ideas. For updates and announcements, keep an eye on our GitHub repo and Ultralytics Docs. Your enthusiasm and suggestions are much appreciated!

Logeshmuthu commented 4 months ago

it's possible to seg, keypoint , oBB modeles are work in ios devices via yolov8 to coreml converted ? @glenn-jocher

glenn-jocher commented 3 months ago

Hello @Logeshmuthu! 😊

Yes, it is possible to run segmentation (seg), keypoint, and oriented bounding box (oBB) models on iOS devices using YOLOv8 models converted to Core ML. The Ultralytics iOS App leverages the Apple Neural Engine and Core ML for optimized performance, making it feasible to deploy these models on iOS devices.

Here's a general outline of the steps you would follow to convert and deploy these models:

  1. Train Your Model: Train your YOLOv8 model for segmentation, keypoint detection, or oriented bounding boxes using the Ultralytics framework.

  2. Export to Core ML: Use the Ultralytics export functionality to convert your trained model to Core ML format. You can do this with the following command:

    yolo export model=yolov8n.pt format=coreml

    Replace yolov8n.pt with the path to your trained model.

  3. Integrate with iOS App: Integrate the Core ML model into your iOS app. You can use the Ultralytics iOS App as a reference for how to set up real-time object detection.

  4. Optimize for Performance: Ensure your model is quantized to FP16 or INT8 for better performance on iOS devices. This can be done during the export process.

If you encounter any issues during this process, please ensure you are using the latest versions of the Ultralytics packages and tools. If the issue persists, feel free to open a new issue with detailed information, and we'll be happy to assist you further.

For more detailed guidance, you can refer to the Ultralytics Docs.

Best of luck with your project, and feel free to reach out if you have any more questions! 🚀