Our codes are based on https://github.com/JacobYuan7/DIN-Group-Activity-Recognition-Benchmark.
I deeply appreciate their efforts. This is the official repository for the following paper:

Chihiro Nakatani, Hiroaki Kawashima, Norimichi Ukita
Learning Group Activity Features Through Person Attribute Prediction, CVPR2024
Project page: https://toyota-ti.ac.jp/Lab/Denshi/iim/ukita/selection/CVPR2024-GAFL.html

python 3.10.2
ROIAlign (https://github.com/longcw/RoIAlign.pytorch)

And you can use requirements.txt

pip install -r requirements.txt

You can download daatset from the following url.
These dataset are required to place in data/ in the repository as follows:

• Volleyball dataset (data/volleyball/videos)
  https://github.com/mostafa-saad/deep-activity-rec

• Collective Activity dataset (data/collective)
  https://cvgl.stanford.edu/projects/collective/collectiveActivity.html

• You can change parameters of the model by editing the files located in scripts (e.g., scripts/train_volleyball_stage2_gr.py).
• Trained model are also published in here (https://drive.google.com/drive/folders/1UnwII6cHG-5SMVPAHwweO92TUwXQfKqt?usp=drive_link).
• trained models required to place in result/ (e.g., result/GAFL_PAC_VOL).

• Ours

python scripts/train_volleyball_stage2_gr.py

The following folder contains the trained models.

1. GAFL_PAC_VOL (GAFL-PAC)
2. GAFL_PAF_VOL (GAFL-PAF)

• Ours

python scripts/train_collective_stage2_gr.py

The following folder contains the trained models.

1. GAFL_PAC_CAD (GAFL-PAC)
2. GAFL_PAF_CAD (GAFL-PAF)

You can choose the model that you would like to evaluate in the bash file script.

• Ours

bash ./evaluation_vol.bash

• Ours

bash ./evaluation_cad.bash