Welcome to "You Can See but Not Capture," a React application designed to detect specific objects, such as cell phones, in real-time using the webcam feed. Built with TensorFlow.js and COCO-SSD, this application aims to enhance privacy and security by identifying and responding to the presence of unauthorized devices.
This application uses machine learning to monitor live video streams for the presence of cell phones, triggering a response when such devices are detected. It's an ideal solution for security-sensitive environments where camera use should be restricted.
- Real-Time Detection: Utilizes TensorFlow.js and COCO-SSD to detect cell phones in real-time through a webcam feed.
- Privacy Protection: Automatically obscures the screen when a cell phone is detected, preventing camera-based information capture.
- Customizable Responses: Easily modify the response actions to fit specific security needs or environments.
- Webcam Integration: Seamlessly integrates with webcams using the React Webcam package.
- Minimalist Interface: Features a clean, user-friendly interface that does not distract from the main functionality.
- React: For building the user interface and managing application state.
- TensorFlow.js: Empowers the application with machine learning capabilities.
- COCO-SSD Model: Used for object detection within the video feed.
- Webcam Component: Captures video input from users' webcams.
To set up the project locally, follow these steps:
- Clone the repository:
git clone https://github.com/satyasn01/you-can-see-but-not-capture.git
- Navigate to the project directory:
cd you-can-see-but-not-capture
- Install dependencies:
npm install
- Start the application:
This will run the app in development mode. Open
npm start
http://localhost:3000
to view it in the browser.
Feel free to fork the repository, make changes, and submit pull requests. You can also open issues for bugs or enhancements you'd like to see.
For further inquiries or to discuss collaboration opportunities:
- Email: [email protected]
- GitHub: satyasn01