ai-image-detector / README.md
Ashish Ranjan Karn
init 2
dd849f5

A newer version of the Gradio SDK is available: 5.42.0

Upgrade
metadata
title: AI Image Detector
emoji: 🤖
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 5.34.2
app_file: app.py
pinned: false
license: mit

🤖 AI Image Detector

Detect whether an image is AI-generated or real using state-of-the-art machine learning models.

Overview

This Gradio app uses a specialized model to classify images as either AI-generated or real. The model has been specifically trained to detect images generated by various AI systems including:

  • DALL-E
  • Midjourney
  • Stable Diffusion (SDXL)
  • And other diffusion models

How to Use

  1. Upload an Image: Click on the upload area or drag and drop an image file (JPG, PNG, etc.)
  2. Get Results: The model will analyze your image and return probability scores
  3. Interpret Results: Higher probability for "AI-generated" suggests the image was created by AI

Model Information

  • Task: Image Classification (Binary: AI-generated vs Real)
  • Framework: Transformers + PyTorch
  • Interface: Gradio

Limitations

⚠️ Important Notes:

  • The model may not be 100% accurate on all images
  • Performance may vary depending on the specific AI model used to generate the image
  • Very high-quality AI images or heavily post-processed real images might be misclassified
  • The model is primarily trained on SDXL-style generated images

Technical Details

The app uses direct model inference to provide robust classification results. The model outputs probabilities for each class, giving you confidence scores for the prediction.

Development

This space is built with:

  • Gradio: For the web interface
  • Transformers: For model loading and inference
  • PyTorch: As the backend framework

This is an educational tool for demonstrating AI image detection capabilities. Always use critical thinking when evaluating image authenticity.