Inspiration
With the rise of deepfake technology, many people are victims of fake photos and videos that harm their reputation and privacy. I wanted to build a simple tool that helps people quickly check whether their media is authentic or manipulated.
What it does
The website allows users to upload an image or a short video. It runs metadata analysis (EXIF data), error-level analysis (ELA), and frame-by-frame checks to estimate if the file has been tampered with or could be a deepfake. Finally, it gives a suspicion score with reasons.
How we built it
- Frontend: HTML, CSS, JavaScript
- Backend: Node.js + Express.js
- Analysis: Python scripts (ELA), FFmpeg for video frame extraction
- Tools: VS Code, GitHub
Challenges we ran into
- Setting up the backend + connecting it with the frontend
- Handling video processing without making the server too heavy
- Time constraints before submission deadline
Accomplishments that we're proud of
- Built a working prototype that detects manipulated media
- Learned how metadata and ELA can be used to spot edits
- Integrated Node.js with Python and FFmpeg for media analysis
What we learned
- Basics of media forensics
- How to serve a full-stack project locally
- Importance of lightweight solutions for hackathons
What's next
- Integrate a machine learning deepfake detection model (CNN/RNN-based)
- Deploy the project on the cloud for public use
- Make the UI cleaner and mobile-friendly
Log in or sign up for Devpost to join the conversation.