from ultralytics import YOLO

# Load model
model = YOLO('best.pt')

# Run inference
results = model.predict('image.jpg')

# Process results
for result in results:
    boxes = result.boxes
    masks = result.masks
    for box in boxes:
        class_id = int(box.cls[0])
        confidence = float(box.conf[0])
        print(f"Detected: {model.names[class_id]} ({confidence:.2f})")

Using Hugging Face Inference API

python

import requests

API_URL = "https://api-inference.huggingface.co/models/YOUR_USERNAME/archivision-yolo"
headers = {"Authorization": f"Bearer {YOUR_HF_TOKEN}"}

def query(image_url):
    response = requests.post(API_URL, headers=headers, json={"inputs": image_url})
    return response.json()

output = query("https://example.com/building.jpg")
print(output)

JavaScript/TypeScript

typescript

async function detectArchitecturalElements(imageUrl: string) {
  const response = await fetch(
    'https://api-inference.huggingface.co/models/YOUR_USERNAME/archivision-yolo',
    {
      method: 'POST',
      headers: {
        'Authorization': `Bearer ${process.env.HUGGINGFACE_TOKEN}`,
        'Content-Type': 'application/json'
      },
      body: JSON.stringify({ inputs: imageUrl })
    }
  );
  
  const predictions = await response.json();
  return predictions;
}

Training

This model was trained on architectural images with the following elements: window, door, column, arch, balcony, roof, facade, stairs, chimney, wall

Limitations

Performance may vary on architectural styles not well-represented in training data
Best results on clear, well-lit building photos
May require confidence threshold tuning for specific use cases

License

MIT License

Citation

If you use this model, please cite:

@misc{archivision-yolo,
  author = {ArchiVision Team},
  title = {ArchiVision YOLO Model},
  year = {2025},
  publisher = {Hugging Face},
  url = {https://huggingface.co/YOUR_USERNAME/archivision-yolo}
}

TanTan2025/archivision-yolo

ArchiVision YOLO Model

Model Details

Classes

Usage

Using Ultralytics