An AI-powered Yoga Trainer that uses computer vision and machine learning to guide users through yoga poses and provide real-time feedback on posture accuracy.
- 🧍♂️ Pose detection using computer vision (e.g., Mediapipe / OpenPose)
- 📷 Real-time webcam support
- ✅ Feedback on correct/incorrect posture
- 🧠 Trained ML models for pose classification
- 📈 Accuracy scoring for each pose
- 📹 Optional recording/playback for reviewing sessions
- Python 🐍
- OpenCV 🎥
- MediaPipe / TensorFlow / PyTorch 🧠
- Streamlit or Tkinter (for UI) 🖥️
git clone https://github.com/your-username/ai-yoga-trainer.git
cd ai-yoga-trainer
2. Install Dependencies
bash
Copy code
pip install -r requirements.txt
3. Run the Application
bash
Copy code
python app.py
Or, if using Streamlit:
bash
Copy code
streamlit run app.py
📂 Project Structure
bash
Copy code
ai-yoga-trainer/
│
├── app.py # Main application script
├── pose_detector.py # Pose detection logic
├── utils/ # Helper functions
├── models/ # Trained ML models
├── dataset/ # Pose images or keypoints
├── README.md # Project description
└── requirements.txt # Python dependencies
🧠 How It Works
Captures webcam video
Detects body keypoints using pose detection models
Compares keypoints with reference poses
Gives real-time feedback (alignment, posture, accuracy score)
🤖 Future Improvements
Add voice guidance for yoga sessions
Support for dynamic flows (like Sun Salutation)
Integration with wearables (e.g., fitness bands)
Personalized training plans
📜 License
This project is licensed under the MIT License. See the LICENSE file for details.
🙌 Acknowledgements
MediaPipe by Google
OpenPose by CMU
Streamlit
📬 Contact
For feedback or contributions, open an issue or reach out at [your-email@example.com].
yaml
Copy code
---
Let me know if you want a custom `.gitignore` file or help generating a `requirements.txt`.