https://github.com/protyayofficial/hypc-net

HYPC-Net combines deep convolutional neural networks with classical machine learning techniques to achieve superior accuracy in classifying yoga poses. This project includes a comprehensive analysis of model performance using the Yoga-82 dataset, offering a comparative study against state-of-the-art CNN models.
https://github.com/protyayofficial/hypc-net

hybrid-transfer-learning image-classification image-processing meta-features yoga-pose-classification

Last synced: about 2 months ago
JSON representation

HYPC-Net combines deep convolutional neural networks with classical machine learning techniques to achieve superior accuracy in classifying yoga poses. This project includes a comprehensive analysis of model performance using the Yoga-82 dataset, offering a comparative study against state-of-the-art CNN models.

• Host: GitHub
• URL: https://github.com/protyayofficial/hypc-net
• Owner: protyayofficial
• License: mit
• Created: 2024-08-15T17:17:16.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-08-28T14:50:39.000Z (4 months ago)
• Last Synced: 2024-08-29T17:36:33.993Z (4 months ago)
• Topics: hybrid-transfer-learning, image-classification, image-processing, meta-features, yoga-pose-classification
• Language: Python
• Homepage:
• Size: 8.79 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# HYPCNet: A Hybrid Yoga Pose Classification Neural Network

This project aims to be a benchmark on classifying yoga poses using pure Convolutional Neural Networks (CNN) and without using any specialized technique. This is achieved by using a novel idea of using ConvNext as a backbon to extract metafeatures from images and using conventional classifying machine learning techniques to correcty predict yoga poses with higher accuracy.

## Directory Structure

```

📦 HYPCNet

├─ train.py

utils.py

├─ yoga82

│  ├─ yoga_train

│  │  ├─ class_6

│  │  ├─ class_20

│  │  └─ class_82

│  └─ yoga_test

│     ├─ class_6

│     ├─ class_20

│     └─ class_82

└─ out

   ├─ models

   ├─ test_{class_name}.csv

   ├─ {model}_{class_name}_training_metrics.csv

   └─ models

      └─ {model}_{class_name}_new_best_model.pth

```

©generated by [Project Tree Generator](https://woochanleee.github.io/project-tree-generator)

## Features

- **Hybrid Model Architecture**: Integration of ConvNeXt with traditional ML models like XGboost, RandomForest etc.

- **Few Shot Like Learning Abilities**: Metafeatures extraction helps in classifying classes with limitations.

- **Extensive Assesment**: Detailed metrics comparisions with other STOA models and contemporary models.

## Download the Dataset

To download the dataset: https://forms.gle/tzVHwzbzCEYzZd9W8

More details about the dataset: https://sites.google.com/view/yoga-82/home

Kindly give proper citation to the original authors

## Acknowledgments

We would like to thank the authors of the Yoga-82 repository for providing a solid foundation for our work. Their initial framework was essential in developing our enhanced model.

## License

This project is licensed under the MIT License.