Automated mask generation for efficient YOLO-based instance segmentation in marine environments for fish detection
Instrumentation viewpoint · 2024-01 · (23) · 19-21
DOI: 10.5821/iwp.2024.23.14134 ↗Kurz gefasst
Fachartikel in der Zeitschrift Instrumentation viewpoint, 2024-01. Verfasst von Xènia Rovira Coll und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
This paper addresses the laborious, time-consuming and error- prone process of generating ground truth data to perform instance segmentation of fish in their natural habitat. Our proposal is to use the Segment Anything Model (SAM), which allows zero- shot inference, to automatically build the segmentation masks, significantly reducing the dataset creation time and enhancing scalability to larger datasets. Experimental results using You Only Look Once (YOLO) demonstrate only marginal performance differences between our approach –with segmentation masks created with no human intervention– and a standard training using a fully human-labeled dataset. The results underscore the effectiveness of the automated workflow discussed herein, showcasing substantial reduction in dataset creation time, particularly in demanding underwater scenarios.
Zitieren
APA
Coll, X. R., & Burguera, A. B. (2024). Automated mask generation for efficient YOLO-based instance segmentation in marine environments for fish detection. Instrumentation viewpoint, 19-21. https://doi.org/10.5821/iwp.2024.23.14134
Vancouver
Coll XR, Burguera AB. Automated mask generation for efficient YOLO-based instance segmentation in marine environments for fish detection. Instrumentation viewpoint. 2024:19-21. doi:10.5821/iwp.2024.23.14134
BibTeX
@article{Coll2024,
author = {Coll, Xènia Rovira and Burguera, Antoni Burguera},
title = {Automated mask generation for efficient YOLO-based instance segmentation in marine environments for fish detection},
journal = {Instrumentation viewpoint},
year = {2024},
number = {23},
pages = {19--21},
doi = {10.5821/iwp.2024.23.14134}
}
EUVIMED