This is the official repository of
MetaCrowd: A Unified Framework for Crowd Counting and Traffic Congestion Detection.
conda env create -f environment.yml
conda activate anomalyFor RGBT-CC dataset, please download it from this link.
For ShanghaiTech RGB-D dataset, please download it from this repo.
For UCF-QNRF dataset, please download it from this link
To use our model, follow the code snippet bellow:
# Train and Test CSCA model
bash train_rgbt_cc.sh
bash test_rgbt_cc.sh
bash train_shanghai_rgbd.sh
bash test_shanghai_rgbd.sh
# Train and Test IADM model
bash train_rgbt_cc.sh
bash test_rgbt_cc.sh
bash train_shanghai_rgbd.sh
bash test_shanghai_rgbd.sh
# Train and Test BayesCrowd model
bash train_ucf.sh
bash test_ucf.sh| Models | RGBT-CC | ShanghaiTech RGB-D | UCF-QNRF |
|---|---|---|---|
| CSCA (ACCV 2022) | ✔️ | ✔️ | ❌ |
| IADM (CVPR 2021) | ✔️ | ✔️ | ❌ |
| BayesCrowd (ICCV 2019) | ❌ | ❌ | ✔️ |
If you find our work useful, please cite the following:
@misc{Chi2023,
author = {Chi Tran},
title = {MetaCrowd: A Unified Framework for Crowd Counting and Traffic Congestion Detection},
publisher = {GitHub},
booktitle = {GitHub repository},
howpublished = {https://github.com/SKKU-AutoLab-VSW/ETSS-07-CongestionDetection},
year = {2024}
}
If you have any questions, feel free to contact Chi Tran
([email protected]).
Our framework is built using multiple open source, thanks for their great contributions.