TraceBack is a full-stack web application built with React and Flask that allows you to:
- Upload an image via drag & drop or file picker
- Detect and compare it with existing images
- See visual similarity scores
- Analyze if it's AI-generated using DeepFace
- Export results to CSV or PDF
- 🔍 Reverse image lookup using OpenCV feature matching (ORB)
- 🧬 Perceptual hashing for fast comparisons
- 🤖 AI image detection using DeepFace
- 📊 Visual similarity indicators
- 📁 Drag and drop upload
- 📤 Export results (CSV / PDF)
| Frontend | Backend |
|---|---|
| React + Axios | Flask |
| HTML5 Drag-and-Drop | OpenCV (ORB) |
| FileSaver.js, jsPDF | DeepFace + Pillow + ImageHash |
git clone https://github.com/Nuraj250/traceback.git cd tracebackcd backend python -m venv venv source venv/bin/activate # or venv\Scripts\activate on Windows pip install -r requirements.txt python app.pyFlask runs on:
http://localhost:5000
Make sure you have test images in backend/static/database/.
cd ../frontend npm install npm startReact runs on:
http://localhost:3000
traceback/ ├── backend/ │ ├── app.py │ ├── static/uploads/ │ ├── static/database/ │ ├── utils/image_search.py ├── frontend/ │ ├── src/App.jsx │ ├── src/components/ResultCard.jsx │ ├── public/ │ └── docs/ │ ├── screenshot-upload.png │ └── screenshot-results.png ├── README.md - 📄 Click “Export CSV” to download results in
.csv - 🧾 Click “Export PDF” for a clean printable report
- Image hosting (Cloudinary / S3)
- Save user history
- Real-time similarity percentage tuning
- Mobile UI optimization
MIT © Nuraj

