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 anomaly

For 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

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.

• chen-judge/RGBTCrowdCounting
• AIM-SKKU/CSCA
• ZhihengCV/Bayesian-Crowd-Counting