New Release: Ultralytics v8.3.29

:tada: Announcing Ultralytics YOLO v8.3.29 Release!

:star2: Summary

We are thrilled to announce the release of Ultralytics YOLO v8.3.29! This update introduces groundbreaking features, including the ability to export YOLOv8 models to the Sony IMX500 format, enhancing AI deployment on devices like Raspberry Pi AI Cameras. Dive into the details below to explore how these advancements can elevate your projects.

:bar_chart: Key Changes

  • Sony IMX500 Export Support: You can now export models in the Sony IMX500 format, a crucial feature for devices with limited resources.
  • New FXModel Class: This class increases compatibility with torch.fx, enabling advanced model manipulations.
  • Updated .gitignore: Now excludes *_imx_model/ directories, which store exported model artifacts.
  • Documentation and Tests: Weโ€™ve added comprehensive documentation and tests to ensure a smooth user experience with the new export functionality.

:dart: Purpose & Impact

  • Enhanced Device Integration: Exporting to Sonyโ€™s IMX500 format allows efficient AI processing on Raspberry Pi AI Cameras, making edge computing more accessible.
  • Improved User Guidance: Updated documentation provides clear steps for leveraging this new feature, helping developers integrate it quickly.
  • Streamlined Development Process: The FXModel class and IMX500 support simplify model deployment, reducing barriers to implementation on edge devices.

Whatโ€™s Changed

New Contributors

Full Changelog: v8.3.28โ€ฆv8.3.29

We invite you to try out the new features in v8.3.29 and share your feedback with us. Your insights are invaluable in helping us improve and innovate. Thank you for being a part of the Ultralytics community! :rocket:

Release URL