YOLO11 by Ultralytics
| Property | Value |
|---|---|
| License | AGPL-3.0 |
| Language Support | 11 languages |
| Author | Ultralytics |
| Framework | PyTorch |
What is YOLO11?
YOLO11 represents the latest evolution in Ultralytics' YOLO series, building upon previous versions with significant improvements in accuracy and versatility. This cutting-edge model supports multiple vision tasks including object detection, instance segmentation, pose estimation, and image classification.
Implementation Details
The model comes in various sizes (from nano to extra large) optimized for different use cases. It requires Python ≥3.8 and PyTorch ≥1.8, and can be easily installed via pip. The architecture maintains YOLO's single-stage detection approach while introducing advanced features for improved performance.
- Multiple model variants (n, s, m, l, x) for different speed-accuracy tradeoffs
- Support for detection, segmentation, classification, and pose estimation
- ONNX and TensorRT export capabilities
- Multi-GPU training support
Core Capabilities
- Object Detection with COCO-trained models achieving up to 54.7 mAP
- Instance Segmentation with up to 43.8 mAP mask accuracy
- Pose Estimation supporting keypoint detection
- Classification on ImageNet with up to 79.5% top-1 accuracy
- Real-time inference capabilities with optimized speed-accuracy balance
Frequently Asked Questions
Q: What makes this model unique?
YOLO11 stands out for its comprehensive multi-language support, improved accuracy across all tasks, and versatile deployment options. It maintains YOLO's signature speed while adding advanced features and improved accuracy.
Q: What are the recommended use cases?
The model is ideal for real-world applications in computer vision, including surveillance, autonomous systems, retail analytics, and industrial inspection. Different model sizes allow deployment from edge devices to high-performance servers.
