ultralytics 8.3.168

v8.3.168 - `ultralytics 8.3.168` Optimize unnecessary native-space calculation (#21379)

on Python PyPI

🌟 Summary

Ultralytics 8.3.168 delivers a major upgrade to how YOLO models export predictions, introduces a unified annotation method, and enhances both documentation and user-facing tools for a smoother, more consistent experience across the platform. 🚀✨

📊 Key Changes

• Unified Prediction Export: Refactored and standardized how all YOLO models (detection, segmentation, pose, OBB, RT-DETR) prepare and export predictions to COCO JSON, ensuring consistent handling of image metadata and scaling.
• Adaptive Annotation Labels: Replaced circle_label and text_label with a single, flexible adaptive_label method for drawing labels (circle or rectangle) on images, simplifying both code and usage.
• Streamlit App Image Support: The Ultralytics YOLO Streamlit app now supports image uploads and inference, letting users run detection on images as well as video and webcam feeds.
• Improved Documentation & Examples: Updated guides for YOLO11 and Sony IMX500 integration, clarified class naming, and improved code readability and examples throughout the docs.
• Scoped Imports for Efficiency: Optimized module imports in the VisualAISearch solution for better efficiency and reliability.
• Accurate Author Attribution: Corrected author information in YOLO11 documentation for proper citation.

🎯 Purpose & Impact

• Consistency & Reliability: All YOLO models now export results in a standardized way, reducing errors and making downstream evaluation (like COCO metrics) more robust.
• Simplified Annotation Workflow: The new adaptive_label method makes it easier for users and developers to annotate images, with less code and more flexibility.
• Broader Accessibility: Streamlit app improvements mean users can quickly test models on images without needing to write code or use the command line.
• Better Developer Experience: Cleaner code, improved documentation, and consistent naming conventions make it easier for both new and experienced users to integrate and extend Ultralytics solutions.
• Professionalism: Accurate citations and clear documentation reinforce Ultralytics’ commitment to quality and community trust.

This release is all about making Ultralytics tools more powerful, intuitive, and accessible for everyone—from researchers and developers to everyday users. 💡🖼️🛠️

What's Changed

• Update Ultralytics YOLO11 docs citation section by @RizwanMunawar in #21407
• Fix YOLO naming for imx500 docs by @ambitious-octopus in #21406
• Unify circle_label and text_label into adaptive_label by @RizwanMunawar in #21377
• Scope text_model module import for Solutions by @Laughing-q in #21436
• Add image inference support to Streamlit application by @RizwanMunawar in #21413
• Fix indentation for docs code example by @Laughing-q in #21441
• ultralytics 8.3.168 Optimize unnecessary native-space calculation by @Laughing-q in #21379

Full Changelog: v8.3.167...v8.3.168