Domain-Variance-and-Modality-Aware-Model-Transfer

This repo is for the paper: VMA: Domain Variance- and Modality-Aware Model Transfer for Fine-Grained Occupant Activity Recognition at IPSN 2022.

To run the demo code, please download the data from this link. Please unzip it first to the project directory and open the run_demo_paper.ipynb. Please run the notebook follows the guidlines inside.

The dataset of the project can be found at Zenodo. Please follow the data pre-process procedure described in the paper if you want to extract the hand-crafted feature. The IMU feature extraction code can be found here. For end-to-end deep feature (which is what I am highly interested in and encourage everyone to try), their is a toy model in the demo_future_direction.ipynb for you to play with.

If you find this code or dataset is usefull, we will be glad if you can cite us in your paper :-)

Recommended Packages:

• Python 3.8+
• Numpy 1.19.5
• Scikit-learn 1.0.1
• Pandas 1.4.0
• Tensorflow 2.8.0

If you are using an Intel chip, you may need this to accelerate the computing:

• scikit-learn-intelex 2021.2.2