Yoga-Pose_Predict

This repository contains code for a Flask web application that predicts yoga poses in images using machine learning models. It includes a YOLO model for keypoint detection and a custom PyTorch model for classification.

Features

• Upload an image of a yoga pose to predict the pose.
• View the predicted pose along with the image.

Installation

1. Clone the repository:

  git clone https://github.com/AyushD95/Yoga-Pose_Predict.git

2. Download Dataset:

As it is not trained you will need to provide data use below kaggele dataset download the dataset and extract it to git cloned folder named Yoga-Pose_Predict extract it in new folder name it as Data .

  https://www.kaggle.com/datasets/tr1gg3rtrash/yoga-posture-dataset

3. Install the required Python packages:

  cd Yoga-Pose_Predict

  pip install -r requirements.txt

4. Run the Flask application:

  python app.py

5. Open a web browser and go to below address use the application

  http://localhost:5000

Usuage

1. Click on "Choose an image" button to select image of a yoga pose.

2. Click on the "Predict Pose" button to predict the pose in the image.

3. The predicted pose will be displayed below the image.

Follow Me on LinkedIn

If you find this project interesting or useful feel free to connect with me on LinkedIn for more updates and projects.