UMD (Unified Maritime Dataset)

Read full text Summarise with VIRAT
Authors: Luca Tarasi
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This dataset accompanies the paper *A Multimodal Dataset for Autonomous Surface Vehicle Situational Awareness*. It aims to collect the marine datasets available for surface obstacle detection and tracking tasks, and provide annotations for them. The annotations are presented in JSON format. An API for accessing and editing the data, as well as reproducing the tests proposed in the paper, is given at https://github.com/GRAAL-Lab/pods/tree/v2.0-isme_robotics/ (branch name: `paper`). Before using the scripts, remember to modify the file `cfg.yaml`, changing the dataset path with the path of the folder you downloaded the dataset in. **Note on annotations.** Apart from the new data (rgb/new, rgb/new_bg), every subset below comes from a previously published dataset. For all of these, the annotations distributed here have been revised and improved with respect to the source datasets. In the following, the mapping from the dataset IDs reported in the paper to the actual dataset folders is reported, together with a link to each original dataset. | ID | Original dataset | Folder(s) | Original source ||---|---|---|---|| RGB1 | SeaShips 7000 | `rgb/SeaShips_scenes` | Shao et al., IEEE TMM 2018, https://doi.org/10.1109/TMM.2018.2865686 || RGB2 | Singapore Maritime Dataset Plus (SMD-Plus) | `rgb/smd_plus` | https://github.com/kjunhwa/Singapore-Maritime-Dataset-Plus || RGB3 | SPSCD | `spscd_scenes` (sequences with obstacles), `spscd_bg_scenes` (sequences without obstacles) | https://doi.org/10.3390/jmse11030578 || RGB4 | Pep Buoys | `rgb/RF_boats` | https://universe.roboflow.com/labor3/pepbuoys-53xn3 || RGB5 | Lentic waters dataset | `rgb/RF_boats` | https://universe.roboflow.com/fredy-pogly/lentic-waters-dataset || RGB6 | Crane finder dataset | `rgb/RF_cranes` | https://universe.roboflow.com/uom-5zqae/crane-finder || RGB7 | Sailboats dataset | `rgb/RF_boats` (non-sequential frames), `rgb/RF_sailboats_scenes` (sequential frames) | https://universe.roboflow.com/labor3/sailboat-b4qzn|| RGB New | New data (this work) | `rgb/new` (sequences with obstacles), `rgb/new_bg` (sequences without obstacles) | — || IR1 | Singapore Maritime Dataset (SMD) | `ir/smd_ir_scenes` | https://sites.google.com/site/dilipprasad/home/singapore-maritime-dataset || IR2 | Raytron infrared ship database | `ir/raytron` (frames with real boats), `raytron_mockup` (frames with only mockup boats) | https://openai.raytrontek.com/apply/E_Sea_shipping.html || MLT1 | MIT Sea Grant marine perception dataset | `mlt` | https://seagrant.mit.edu/auvlab-datasets-marine-perception-2-3/ |

Fields of Study & Subjects

Maritime Navigation and Safety Autonomous Vehicle Technology and Safety Advanced Neural Network Applications

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
26 March 2026
Language
EN
Document type
Dataset