A new article, “𝐀 𝐃𝐚𝐭𝐚𝐬𝐞𝐭 𝐟𝐨𝐫 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐒𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐔𝐧𝐝𝐞𝐫𝐰𝐚𝐭𝐞𝐫 𝐌𝐚𝐫𝐢𝐧𝐞 𝐃𝐞𝐛𝐫𝐢𝐬 𝐢𝐧 𝐒𝐡𝐚𝐥𝐥𝐨𝐰 𝐖𝐚𝐭𝐞𝐫𝐬” by Antun Đuraš, Ben J. Wolf, Athina Ilioudi, Ivana Palunko & Bart De Schutter, was published in Scientific Data journal on August 24th 2004.

This paper introduces the Seaclear Marine Debris Dataset, the first publicly available shallow-water marine debris dataset annotated for instance segmentation/object detection. The dataset contains 8610 images collected using ROVs at multiple locations and with different cameras, annotated for 40 object categories, encompassing not only litter but also observed animals, plants, and robot parts. As part of the technical validation, baseline results are provided for object detection using Faster RCNN and YOLOv6 models. Furthermore, the non-triviality of generalizing the trained model performance to unseen sites and cameras due to domain shift is demonstrated. This underscores the value of the presented dataset in further developing robust models for underwater debris detection.

The article is available in open access at https://doi.org/10.1038/S41597-024-03759-2 and can be cited using its DOI: 10.1038/S41597-024-03759-2