DogWatch AI

A Python based Streamlit web app to detect AI-generated images

Table of Contents

• Description
• Features
• Technologies Used
• Setup Instructions
• Deployment

Description

This Streamlit web app utilizes a custom-built convolutional neural network (CNN) that we developed and trained specifically for 10 distinct categories. The model analyzes image arrays to detect AI-generated content, providing detailed scores for each image.

Groups

1. Airplane
2. Automobile
3. Bird
4. Cat
5. Deer
6. Dog
7. Frog
8. Horse
9. Ship
10. Truck

Features

• Detects AI-generated content in images
• Generate AI-images based on user prompt to test app
• Guessing game to interact with users

Technologies-used

• Streamlit: To build to interactive web app.
• Python: Programming language.
• TensorFlow: Deep learning fask and image processing
• NumPy: Arrays for machine learning
• PyTorch:
• OS Module: Used to interact with file paths
• PIL: Image processing
• Hugging Face Hub: Pre-trained machine learning model and API

Setup Instructions

---

Deployment

---