/traditional-dance-video-classification

Balinese traditional dance video classification - Research Project

Primary LanguageJupyter NotebookMIT LicenseMIT

Balinese Dance Classification

An AI project that use machine learning to classify traditional Balinese dance. We used pretrained model VGG-LSTM and I3D. If you're interested in this project please give star and fork our repository. If you want to use our project for your research please cite our paper at JANAPATI (Indonesian language)

Prerequisites

First-thing-first is to clone the repository by following the command below

git clone https://github.com/rudyhendrawn/traditional-dance-video-classification.git
cd traditional-dance-video-classification # go to newly cloned directory

Then, the next step after cloning the repository is to install the neccesary dependencies. You can do that by following the command below

# auto installing from the requirements.txt file
pip install --no-cache-dir -r requirements.txt

# individually install dependencies
pip install ipython
pip install matplotlib
pip install numpy
pip install pandas
pip install scikit-learn
pip install seaborn
pip install tensorflow

You can make sure to create venv first to install the dependencies locally and not globally. Virtual environment can be created using this command: python3 -m venv ./venv or python -m venv ./venv. Do the activation using the command: linux/macos source venv/bin/activate or windows .\venv\Scripts\activate. If not, please make sure that you know what you're doing.

CUDA (Optional)

You can do the CUDA installation if neccesary/needed. You can download the CUDA Toolkit here and cuDNN here. Please refer to this video for installation tutorial.

Usage

For the usage, you can install some extension in your VS Code for Python and Jupyter Notebook. Then, run it from there!

  • Google Colab compatibility: Coming Soon
  • Paperspace Gradient compatibility: Coming Soon

DOI

If you use our code, please cite our code.

DOI