YOLOv8m PPE / Helmet Detection
A YOLOv8-medium model fine-tuned to detect Personal Protective Equipment (PPE) in images and video.
Classes (14)
| ID | Class | ID | Class |
|---|---|---|---|
| 0 | Fall-Detected | 7 | NO-Goggles |
| 1 | Gloves | 8 | NO-Hardhat |
| 2 | Goggles | 9 | NO-Mask |
| 3 | Hardhat | 10 | NO-Safety Vest |
| 4 | Ladder | 11 | Person |
| 5 | Mask | 12 | Safety Cone |
| 6 | NO-Gloves | 13 | Safety Vest |
Quick Start
from ultralytics import YOLO
# Load from Hugging Face
model = YOLO("hf://Hexmon/vyra-yolo-ppe-detection/best.pt")
# Run inference
results = model.predict("image.jpg", conf=0.5)
results[0].show()
Training Details
- Base model: YOLOv8m (pretrained on COCO)
- Image size: 640
- Epochs: 100
- Batch size: 32
- Dataset: PPE Combined Model v4 (CC BY 4.0)
Usage Examples
Webcam:
model.predict(source=0, show=True, conf=0.5)
Video:
model.predict("video.mp4", conf=0.5, save=True)
Access detections:
results = model.predict("image.jpg", conf=0.5)
for box in results[0].boxes:
print(model.names[int(box.cls)], float(box.conf))