Camera-based Intruder Detection and Monitoring of Ship Crew Work Hours

---

@InProceedings{Murad_2025_WACV,
    author    = {Murad, Md Mahmuddun Nabi and Turgut, Bora San and Ahmed, Awwab and Camliyurt, Gokhan and Yilmaz, Yasin},
    title     = {Camera-based Intruder Detection and Monitoring of Ship Crew Work Hours},
    booktitle = {Proceedings of the Winter Conference on Applications of Computer Vision (WACV) Workshops},
    month     = {February},
    year      = {2025},
    pages     = {1526-1534}
}

Get started

Follow these steps to get started:

1. Install Requirements

Install Python 3.10 and the necessary dependencies.

pip install -r requirements.txt

2. Download Data

Download the datasets from the link. Paste the dataset inside ./data/videos/. For example, you will get many sub-folders titled by dirxxxxx in the link. Keep these sub-directories inside the ./data/videos/ in such a way, so that you get ./data/videos/dirxxxxx sub-directories.

3. Experiment

Run main.py file for the experiments. There are three type of tasks for the experiment: 'face_detection', 'crew_recognition', and 'work_hour_estimation'.

'face_detection':

It is used to detect and crop the faces from the video video dataset. Then we manually labeled each faces with a crew id. You do not need to do this experiment, as we already provided these files in ./data/face_photos/unsplit_data.

'crew_recognition':

Experiment to recognize the crew members and to detect the intruders.

'work_hour_estimation':

Experiment for estimating the work hours of the crew memebers.

Contact

For more information: Please contact [email protected]