Motion tracking avatar
Emojiman is a fun project that overlays emojis and images on a person's face and body in real-time using computer vision techniques.
This project uses the Mediapipe library along with OpenCV to detect and track the pose of a person from a live video feed. It then overlays emojis and images onto specific body parts detected in the video stream.
- Real-time pose detection and tracking.
- Overlay emojis on the face.
- Overlay images of hands and boots on corresponding body parts.
- Detect when a cigarette is close to the face and display smoke effect.
To use this project, make sure you have Python installed along with the necessary dependencies. You can install the required packages by running:
pip install -r requirements.txt
Then, run the emojiman.py script:
python emojiman.py
Press 'q' to exit the application.
- OpenCV
- Mediapipe
- Implement emoji scaling based on shoulder width.
- Refactor the emoji overlay code into a function that takes specific emojis as parameters.
- Add an option to overlay on recorded video.
This project is licensed under the MIT License. See the LICENSE file for details.