Find research datasets worth reusing

MOBDrone: a large-scale drone-view dataset for man overboard detection

<p><strong>Dataset</strong></p> <p>The <em>Man OverBoard Drone (MOBDrone)</em> dataset is a large-scale collection of aerial footage images. It contains 126,170 frames extracted from 66 video clips gathered from one UAV flying at an altitude of 10 to 60 meters above the mean sea level. Images are manually annotated with more than 180K bounding boxes localizing objects belonging to 5 categories --- <em>person, boat, lifebuoy, surfboard, wood</em>. More than 113K of these bounding boxes belong to the person category and localize people in the water simulating the need to be rescued.</p> <p>In this repository, we provide:</p> <ul> <li> <p>66 Full HD video clips (total size: 5.5 GB)&nbsp;</p> </li> <li> <p>126,170 images extracted from the videos at a rate of 30 FPS (total size: 243 GB)</p> </li> <li> <p>3 annotation files for the extracted images that follow the MS COCO data format (for more info see <a href="https://cocodataset.org/#format-data">https://cocodataset.org/#format-data</a>):</p> <ul> <li> <p><em>annotations_5_custom_classes.json</em>: this file contains annotations concerning all five categories; please note that class ids do not correspond with the ones provided by the MS COCO standard since we account for two new classes not previously considered in the MS COCO dataset --- <em>lifebuoy </em>and <em>wood</em></p> </li> <li> <p><em>annotations_3_coco_classes.json:</em> this file contains annotations concerning the three classes also accounted by the MS COCO dataset --- <em>person, boat, surfboard</em>. Class ids correspond with the ones provided by the MS COCO standard.</p> </li> <li> <p><em>annotations_person_coco_classes.json</em>: this file contains annotations concerning only the &#39;<em>person</em>&#39; class. Class id corresponds to the one provided by the MS COCO standard.</p> </li> </ul> </li> </ul> <p>The MOBDrone dataset is intended as a test data benchmark. However, for researchers interested in using our data also for training purposes, we provide training and test splits:</p> <ul> <li><em>Test set: </em>All the images whose filename starts with &quot;DJI_0804&quot; (total: 37,604 images)</li> <li><em>Training set:</em> All the images whose filename starts with &quot;DJI_0915&quot; (total: 88,568 images)</li> </ul> <p>More details about data generation and the evaluation protocol can be found at our MOBDrone paper: <a href="https://arxiv.org/abs/2203.07973">https://arxiv.org/abs/2203.07973</a><br> The code to reproduce our results is available at this GitHub Repository: <a href="https://github.com/ciampluca/MOBDrone_eval">https://github.com/ciampluca/MOBDrone_eval</a><br> See also&nbsp;<strong>&nbsp;</strong><a href="http://aimh.isti.cnr.it/dataset/MOBDrone">http://aimh.isti.cnr.it/dataset/MOBDrone</a></p> <p><strong>Citing the MOBDrone</strong></p> <p>The MOBDrone is released under a Creative Commons Attribution license, so please cite the MOBDrone if it is used in your work in any form.<br> Published academic papers should use the academic paper citation for our MOBDrone paper, where we evaluated several pre-trained state-of-the-art object detectors focusing on the detection of the overboard people</p> <blockquote> <pre>@inproceedings{MOBDrone2021, title={MOBDrone: a Drone Video Dataset for Man OverBoard Rescue}, author={Donato Cafarelli and Luca Ciampi and Lucia Vadicamo and Claudio Gennaro and Andrea Berton and Marco Paterni and Chiara Benvenuti and Mirko Passera and Fabrizio Falchi}, booktitle={ICIAP2021: 21th International Conference on Image Analysis and Processing}, year={2021} } </pre> </blockquote> <p>and this&nbsp;Zenodo Dataset</p> <blockquote> <pre>@dataset{donato_cafarelli_2022_5996890, author={Donato Cafarelli and Luca Ciampi and Lucia Vadicamo and Claudio Gennaro and Andrea Berton and Marco Paterni and Chiara Benvenuti and Mirko Passera and Fabrizio Falchi}, title = {{MOBDrone: a large-scale drone-view dataset for man overboard detection}}, month = feb, year = 2022, publisher = {Zenodo}, version = {1.0.0}, doi = {10.5281/zenodo.5996890}, url = {https://doi.org/10.5281/zenodo.5996890} }</pre> </blockquote> <p>Personal works, such as machine learning projects/blog posts, should provide a URL to the MOBDrone Zenodo page (<a href="https://doi.org/10.5281/zenodo.5996890">https://doi.org/10.5281/zenodo.5996890</a>), though a reference to our MOBDrone paper would also be appreciated.</p> <p>&nbsp;</p> <p><strong>Contact Information</strong></p> <p>If you would like further information about the MOBDrone or if you experience any issues downloading files, please contact us at <a href="mailto:mobdrone@isti.cnr.it">mobdrone[at]isti.cnr.it</a></p> <p>&nbsp;</p> <p><strong>Acknowledgements</strong></p> <p>This work was partially supported by NAUSICAA - &quot;NAUtical Safety by means of Integrated Computer-Assistance Appliances 4.0&quot; project funded by the Tuscany region (CUP D44E20003410009). The data collection was carried out with the collaboration of the Fly&amp;Sense Service of the CNR of Pisa - for the flight operations of remotely piloted aerial systems - and of the Institute of Clinical Physiology (IFC) of the CNR - for the water immersion operations.&nbsp;</p>