Downloads · 30 days
35
30% of all-time downloads
pothole-detection-yolov8
peterhdd/pothole-detection-yolov8
pothole-detection-yolov8 is a object detection model from peterhdd. Use it when you need objects located in an image. The card lists the license as apache-2.0.
This model detects potholes in road images using YOLOv8s trained on the following dataset:
All-time downloads
116
Public
Repo size
68.3 MB
Likes
4
Public
Repo makeup
Click a slice to open those files.
.onnx44.7 MB · 65%
At a glance
- Task
- Object Detection
- License
- apache-2.0
- Model type
- yolo
- Access
- Public
- Created
- Dec 13, 2025
- Updated
- Jan 22, 2026
- SHA
- da7747ee
Base models
Model card
From the Hugging Face model README
YOLOv8 Pothole Detection Model
This model detects potholes in road images using YOLOv8s trained on the following dataset:
Model Details
- Architecture: YOLOv8s (Ultralytics)
- Task: Object Detection
- Classes: 1 (pothole)
- Epochs: 100
- Hardware: NVIDIA L40S (Nebius Cloud)
- Model Size: 22.5 MB
Sample Prediction
Training Results
Confusion Matrix
Usage Example (Python)
from ultralytics import YOLO
model = YOLO("https://huggingface.co/peterhdd/pothole-detection-yolov8/resolve/main/best.pt")
results = model("your_image.jpg")
results.show()
This model is part of a complete end-to-end pothole detection system including training, GPU inference, and a mobile application.
The repository can be found here:
- Task
- Object Detection
- Type
- yolo
- License
- apache-2.0
- Created
- Dec 13, 2025
- Updated
- Jan 22, 2026