Building Detection Technologies to Combat Synthetic Media
AIDA (AI Detection App) is a mobile application that allows users to scan physical artwork and detect whether it was generated by AI. The app utilizes a U²-Net for object isolation, a custom-trained CNN for classification, and Grad-CAM to visually explain results. Unlike typical detectors, AIDA works on second-order images (e.g., photos of prints or posters) that lack watermarks or metadata.
-
Clone the repostory in a new folder.
-
cd into the dist folder cd dist
-
Have Node.js installed and install expo npm install -g expo-cli
-
run npm install, and npm update if there are any outdated packages. npm install npm update
-
start the app. Download any necessary packages. npx expo start
-
Download Expo Go on your iPhone and scan the QR code to simulate the app https://apps.apple.com/us/app/expo-go/
-
create a .env file with the following tokens: DATABASE_URL AWS_ACCESS_KEY_ID AWS_SECRET_ACCESS_KEY AWS_REGION AWS_S3_BUCKET HF_BEARER_TOKEN
Contributors: Claire Pielak, Zachary Lain, Chris Stefaniak, Luke Renchik, Savannah Moss
Institution: Department of Computer Science and Engineering School of Engineering and Computer Science Oakland University, Rochester Hills, MI
License MIT License