The datasets are acceccable through following link:
MaSTr1325 Images [512x384]
MaSTr1325 Ground Truth Annotations [512x384]
Reference:
@inproceedings{bb_iros_2019,
title={The MaSTr1325 dataset for training deep USV obstacle detection models},
author={Bovcon, Borja and Muhovi{\v{c}}, Jon and Per{\v{s}}, Janez and Kristan, Matej},
booktitle={2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
year={2019},
organization={IEEE}
}
Ocean AI Segmentation Initiatives (OASIs) dataset
Reference:
@article{kim2024introducing,
title={Introducing VaDA: Novel Image Segmentation Model for Maritime Object Segmentation Using New Dataset},
author={Kim, Yongjin and Park, Jinbum and Kang, Sanha and Kim, Hanguen},
journal={arXiv preprint arXiv:2407.09005},
year={2024}
}