Inspiration
Streamers often accidentally reveal credit cards on camera, risking identity theft and financial fraud. We wanted to create a tool that protects creators in real-time so they can stream safely.
What it does
CD Blocker detects and blurs credit cards in live streams before the video reaches platforms like YouTube or Twitch, ensuring sensitive information never gets exposed.
How we built it
We combined computer vision with Gemini to detect and blur sensitive items. The system processes video frames in real-time and applies a blur over sensitive areas with minimal latency.
Challenges we ran into
Running out of Gemini daily free Ensuring blur happens fast enough not to disrupt the stream
Accomplishments that we're proud of
A fully working prototype that integrates seamlessly with streaming platforms We used technology we never used before Learned more about detection and Computer Vision
What we learned
Real-time video processing requires careful optimization Combining AI models with traditional validation improves reliability
What's next for CD Blocker
Connection with Youtube and an AI Agent that calls the user if it detects to get a better response from the user to increase safety
Built With
- gemini
- opencv
- python
- react
- twailand
- typescript
- vite
- vscode
- websocket
Log in or sign up for Devpost to join the conversation.