AI Image Detector

Model Description

This model is designed to detect whether an image is real or AI-generated. It uses Vision Transformer (ViT) architecture to provide accurate classification.

Model Usage

from transformers import ViTImageProcessor, ViTForImageClassification
from PIL import Image
import torch

# تحميل النموذج والمعالج
processor = ViTImageProcessor.from_pretrained("C:/Users/SUPREME TECH/Desktop/SAM3/ai-image-detector")
model = ViTForImageClassification.from_pretrained("C:/Users/SUPREME TECH/Desktop/SAM3/ai-image-detector")

def detect_image(image_path):
    # فتح وتجهيز الصورة
    image = Image.open(image_path)
    if image.mode != 'RGB':
        image = image.convert('RGB')
    
    # معالجة الصورة
    inputs = processor(images=image, return_tensors="pt")
    
    # الحصول على التنبؤات
    with torch.no_grad():
        outputs = model(**inputs)
        predictions = outputs.logits.softmax(dim=-1)
    
    # تحليل النتائج
    scores = predictions[0].tolist()
    results = [
        {"label": "REAL", "score": scores[0]},
        {"label": "FAKE", "score": scores[1]}
    ]
    
    # ترتيب النتائج حسب درجة الثقة
    results.sort(key=lambda x: x["score"], reverse=True)
    
    return {
        "prediction": results[0]["label"],
        "confidence": f"{results[0]['score']*100:.2f}%",
        "detailed_scores": [
            f"{r['label']}: {r['score']*100:.2f}%" 
            for r in results
        ]
    }

# كود للاختبار
if __name__ == "__main__":
    # يمكنك تغيير مسار الصورة هنا
    image_path = "path/to/your/image.jpg"
    
    try:
        result = detect_image(image_path)
        print("\nنتائج تحليل الصورة:")
        print(f"التصنيف: {result['prediction']}")
        print(f"درجة الثقة: {result['confidence']}")
        print("\nالتفاصيل:")
        for score in result['detailed_scores']:
            print(f"- {score}")
            
    except Exception as e:
        print(f"حدث خطأ: {str(e)}")

Classes

The model classifies images into two categories:

  • Real Image (0): The image is real and not AI-generated.
  • AI Generated (1): The image is generated by AI.

Technical Details

  • Model Architecture: Vision Transformer (ViT)
  • Input: Images (RGB)
  • Output: Binary classification with confidence score
  • Max Image Size: 224x224 (automatically resized)

Requirements

  • transformers>=4.30.0
  • torch>=2.0.0
  • Pillow>=9.0.0

Limitations

  • Best performance with clear, high-quality images.
  • May have reduced accuracy with heavily edited photos.
  • Designed for general image detection.

Web Integration Example

async function detectImage(imageFile) {
  const formData = new FormData();
  formData.append('image', imageFile);

  const response = await fetch('YOUR_API_ENDPOINT', {
    method: 'POST',
    body: formData
  });

  return await response.json();
}

Developer

Downloads last month
196
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.