Human Action Recognition Using IMU and Photogrammetry Data

This repository contains the data used in the paper Accuracy Comparison of CNN, LSTM, and Transformer for Activity Recognition Using IMU and Visual Markers, including the notebooks used for training and testing.

Datasets

The datasets can be accessed by the following public Google Drive link: https://drive.google.com/drive/folders/1k2sAkmRyyctE1uOc19mrixyt2N47-7pt?usp=sharing

Note: In order to use the notebooks provided, you need to make a copy or a shortcut of the folder in the root of your Google Drive (My Drive).

Notebooks

The notebooks are divided according to their dataset and the HAR technique (IMU or visual markers). Inside the Notebooks folder, there are 3 subfolders, corresponding to each dataset used, for clarity and ease of access.

Credits

• María Fernanda Trujillo (Collection of the datasets)
• Stadyn Román Niemes (Coding and creation of the notebooks)
• Ricardo Fonseca (Technical help)

Citation

Paper available at: https://doi.org/10.1109/ACCESS.2023.3318563

If you use our data, code, or paper for your research, please include the appropriate citation in your article:

@ARTICLE{10261772,
  author={Trujillo-Guerrero, María Fernanda and Román-Niemes, Stadyn and Jaén-Vargas, Milagros and Cadiz, Alfonso and Fonseca, Ricardo and Serrano-Olmedo, José Javier},
  journal={IEEE Access}, 
  title={Accuracy Comparison of CNN, LSTM, and Transformer for Activity Recognition Using IMU and Visual Markers}, 
  year={2023},
  volume={11},
  number={},
  pages={106650-106669},
  doi={10.1109/ACCESS.2023.3318563}}