Ultralytics YOLO

🌟 Summary

This release (v8.3.42) focuses on fixing computation issues, improving inference flexibility, simplifying user experience, and ensuring clarity in documentation and outputs. 🚀

📊 Key Changes

• Fixed an Array Conversion Bug 🐛: Resolved issues in the np.empty_like function where integer input types caused computation errors in xyxy2xywhn. Now, results are correctly returned as floating-point arrays.
• Batch Inference Added 🎥: Introduced a batch parameter for the predict() function, allowing users to enhance performance on videos, directories, or .txt inputs.
• Segmentation Output Improved ✂️: Adjusted class name computation in the AutoBackend module for better handling of model output shapes during segmentation tasks.
• Examples for Heatmap & Queue Solutions 📚: Simplified and clarified example codes for Heatmap and QueueManager solutions, making them easier to use for developers.
• Documentation Updates ✏️:
  • Dynamically updated benchmark version placeholders for easier maintenance in Raspberry Pi guides.
  • Improved accuracy in links and navigation (e.g., fixed Discord and homepage links).
  • Added a video tutorial for YOLO11 integration in Google Colab to the official tutorial notebook.

🎯 Purpose & Impact

• Bug Fixes:

  • 🛠️ Purpose: Ensure accurate bounding box transformations in common utility functions.
  • 🔍 Impact: Fixes a critical issue with erroneous array initialization that previously resulted in incorrect detections.

• Efficiency & Speed:

  • 🛠️ Purpose: Batch inference boosts throughput for large datasets or long video streams.
  • ⚡ Impact: Reduces processing time, enabling more scalable and efficient workflows on powerful hardware.

• Segmentation Workflow:

  • 🛠️ Purpose: Refined class name handling in segmentation tasks leads to fewer errors in edge cases.
  • ✨ Impact: Users working with segmentation models benefit from a smoother, error-free experience.

• Developer Support:

  • 🛠️ Purpose: More intuitive and beginner-friendly examples for heatmaps and queue management solutions.
  • 🌱 Impact: Simplifies development by aligning examples with practical, real-world scenarios.

• Documentation & Navigation:

  • 🛠️ Purpose: Improve ease of access to key information like benchmarks, tutorials, and community resources.
  • 🌍 Impact: Streamlined guides and corrected links help users onboard and locate critical information faster.

This update empowers users with greater reliability, performance, and clarity, addressing key pain points in both core utilities and the broader developer experience. 🚀

What's Changed

• Update mkdocs.yml homepage by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/17924
• Add batch to list of predict() arguments by @Y-T-G in https://github.com/ultralytics/ultralytics/pull/17979
• Add https://youtu.be/ZN3nRZT7b24 to tutorial.ipynb by @RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/17875
• Improve docs for Ultralytics version in benchmarks by @lakshanthad in https://github.com/ultralytics/ultralytics/pull/17925
• ultralytics 8.3.42 update AutoBackend names placeholder by @Laughing-q in https://github.com/ultralytics/ultralytics/pull/17970
• Update CLI message by @RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/18041
• Update functions description by @RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/18050
• Fix np.empty_like function input type by @RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/18049

Full Changelog: https://github.com/ultralytics/ultralytics/compare/v8.3.40...v8.3.42