🌟 Summary
v8.4.201 makes PIL-based plotting up to 15× faster while keeping visuals consistent, and standardizes the Construction-PPE dataset’s name and download paths. 🚀
📊 Key Changes
- Faster plotting across tasks: Training and validation mosaics and
Results.plot(pil=True)benefit from cached fonts and label patches, fewer image conversions, and more efficient mask blending. - More efficient image handling: Only plotted images are converted for mosaics, and tensor images are transferred as
uint8to reduce overhead. - Construction-PPE naming and provenance updated: Documentation and configuration now use the canonical
construction-ppearchive and directory. The dataset is described as an original Ultralytics dataset, created and open-sourced in November 2026, with 1,000 images, 12 classes, and a CC BY 4.0 license. - No model architecture changes are included in this release.
🎯 Purpose & Impact
- Quicker visual checks: Faster plotting can reduce waiting during training, validation, and prediction review, especially when generating many annotated images.
- Clearer dataset setup: Consistent naming makes it easier to find and use Construction-PPE. Existing local caches using the old directory name are outside the supported scope of this release.
- Same plotting appearance: The optimizations target processing overhead rather than changing how annotations look. ✨
What's Changed
- Correct Construction-PPE naming and provenance by @glenn-jocher in #27015
ultralytics 8.4.201Up to 15x faster PIL plotting across all tasks by @glenn-jocher in #27014
Full Changelog: v8.4.200...v8.4.201