Real-time ai based yoga pose detection using CNN and YOLO
demo.mp4
P.S. I'm really bad at yoga. :)
This project uses YOLOv8 for real-time object detection and a TensorFlow model for yoga pose classification. It captures live video, detects the presence of a person, extracts and analyzes their pose to provide accurate yoga pose identification. This system helps ensure correct yoga practice by providing immediate feedback on pose accuracy.
The datataset used is the roboflow's Yoga Pose Computer Vision Project, which includes numerous images for each of the 107 yoga poses.
Dataset Source Link: data url
-
After downloading the dataset, unzip it and place in the root directory.
-
Clone the repository
git clone https://github.com/priyanshudutta04/Yoga-Pose-Detection.git
- Install dependencies
pip install -r requirements.txt
- Run the Python File
python yoga-detection.py
Note: If GPU is available install cuda toolkit and cuDNN for faster execution during Model-Training
Contributions are welcome! If you have ideas for improving the model or adding new features, please feel free to fork the repository and submit a pull request.
If you like this project, do give it a ⭐and share it with your friends