Seaclear Marine Debris Detection & Segmentation Dataset

doi: 10.4121/4f1dff25-e157-4399-a5d4-478055461689.v1
The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
doi: 10.4121/4f1dff25-e157-4399-a5d4-478055461689
Datacite citation style:
Đuraš, Antun; Ilioudi, Athina; Wolf, Ben; Palunko, Ivana; De Schutter, Bart (2024): Seaclear Marine Debris Detection & Segmentation Dataset. Version 1. 4TU.ResearchData. dataset.
Other citation styles (APA, Harvard, MLA, Vancouver, Chicago, IEEE) available at Datacite

Seaclear Marine Debris Dataset contains 8610 images of marine litter annotated for tasks of object detection and instance segmentation with 40 object categories, encompassing not only litter but also observed animals, plants, and robot parts. Data was captured using ROVs during field experiments related to Seaclear project and thus includes images from multiple locations:

  • Bistrina, Croatia
  • Jakljan, Croatia
  • Lokrum, Croatia
  • Slano, Croatia
  • Marseille, France

Annotations are available as a COCO format (.json) file, and the images are arranged in folders, each dedicated to a unique site-camera pair. All images are resized to 1920x1080 resolution.

  • 2024-01-08 first online, published, posted
  • SEarch, identificAtion and Collection of marine Litter with Autonomous Robots (SeaClear) (grant code 871295) [more info...] European Union’s Horizon 2020 Research and Innovation Programme
Laboratory for Intelligent Autonomous Systems (LARIAT), Department of Electrical Engineering and Computing, University of Dubrovnik

Delft Center for Systems and Control, Delft University of Technology


files (1)