Data downloader and PyTorch dataloader are uploaded in the jupyter notebook. The model weights and the testing code are being released.

This repository contains a notebook that downloads the benchmark dataset used in our paper.

Shah, A., Thomas, L., & Maskey, M. (2021) "Marine Debris Dataset for Object Detection in Planetscope Imagery", Version 1.0, Radiant MLHub. https://doi.org/10.34911/rdnt.9r6ekg

Emanuele Dalsasso, Marc Rußwurm, Christian Donner, Samuel Darmon, Robin de Vries, Michele Volpi, Devis Tuia, Self-supervised pre-training enables marine debris detection across sensors, Remote Sensing of Environment, Volume 339, 2026, 115391, ISSN 0034-4257, https://doi.org/10.1016/j.rse.2026.115391.

@article{DALSASSO2026115391,
title = {Self-supervised pre-training enables marine debris detection across sensors},
journal = {Remote Sensing of Environment},
volume = {339},
pages = {115391},
year = {2026},
issn = {0034-4257},
doi = {https://doi.org/10.1016/j.rse.2026.115391},
url = {https://www.sciencedirect.com/science/article/pii/S0034425726001616},
author = {Emanuele Dalsasso and Marc Rußwurm and Christian Donner and Samuel Darmon and Robin {de Vries} and Michele Volpi and Devis Tuia},
}