Loading repository data…
Loading repository data…
linh-gist / repository
This is the official Python and C++ implementation repository for a paper entitled "Track Initialization and Re-Identification for 3D Multi-View Multi-Object Tracking", Information Fusion (http://arxiv.org/abs/2405.18606).
A transparent discovery signal based on current public GitHub metadata.
This score does not audit code, security, maintainers, documentation quality, or suitability. Verify the repository and its current documentation before adoption.
This is the official Python and C++ implementation repository for a paper entitled "Track Initialization and Re-Identification for 3D Multi-View Multi-Object Tracking", Information Fusion (arXiv 2405.18606).
Docker image to run MV-GLMB-AB codes can be found in Docker Hub 3D-Visual-MOT with YouTube Demo or Docker Hub ISPLCurtin.
Set Up Python Environment
conda Python environment and activate it:
conda create --name virtualenv python==3.7.16
conda activate virtualenv
git clone --recursive https://github.com/linh-gist/3D-Visual-MOT.git
Install Packages
Eigen 3.4.0 is installed): Navigate to the cpp_ms_glmb_ukf folder and run python setup.py build developpip install -r requirements.txt or using conda install -c conda-forge motmetrics lap h5py matplotlib opencvConfigure Experiment Options
adaptive_birth=3, use_feat=True to run experiments with Meanshift Adaptive Birth Methods and to toggle the use of re-identification features.detection_aka_occlusion_model_v2(...), but it can be changed to detection_aka_occlusion_model(...).demo.py uses glmb.runcpp(model_params, dataset, meas, adaptive_birth, use_feat) to run experiments with C++. (Note: make sure to comment this line glmb.run(model_params, dataset, meas)Prepare Data
|-- data
| |-- images
| | |-- CMC1
| | | |-- Cam_1
| | | |-- Cam_2
| | | |-- Cam_3
| | | |-- Cam_4
| | |-- ...
| | |-- CMC5
| | |-- WILDTRACK
|-- source code
| |-- cpp_ms_glmb_ukf
| |-- detection
| | |-- cstrack
| | | |-- CMC1
| | | | |-- Cam_1.npz
| | | | |-- Cam_2.npz
| | | | |-- Cam_3.npz
| | | | |-- Cam_4.npz
| | | |-- CMC2
| | |-- fairmot
| |-- experiments
| |-- ms_glmb_ukf
|-- README.md
../detection/fairmot/ in gen_meas.py.
../../data/images/ in gen_meas.py.gt_data_dir="../../data/images/" for performance evaluation using CLEAR MOT in clearmot.py and OSPA2 in ospa2.py.ospa2.py.
@article{rezatofighi2020trustworthy,
title={How trustworthy are the existing performance evaluations for basic vision tasks?},
author={Tran Thien Dat Nguyen and Hamid Rezatofighi and Ba-Ngu Vo and Ba-Tuong Vo and Silvio Savarese and Ian Reid},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
year={2022}
}
Run the Tracking Demo
ms_glmb_ukf and run python demo.pyLinh Ma (linh.mavan@gm.gist.ac.kr), Machine Learning & Vision Laboratory, GIST, South Korea
If you find this project useful in your research, please consider citing by:
@article{linh2024inffus,
title={Track Initialization and Re-Identification for {3D} Multi-View Multi-Object Tracking},
author={Linh Van Ma, Tran Thien Dat Nguyen, Ba-Ngu Vo, Hyunsung Jang, Moongu Jeon},
journal={Information Fusion},
volume = {111},
year={2024},
publisher={Elsevier}
}