The Marine Debris Dataset for Forward-Looking Sonar Semantic Segmentation
Abstract
Accurate detection and segmentation of marine debris is important for keeping the water bodies clean. This paper presents a novel dataset for marine debris segmentation collected using a Forward Looking Sonar (FLS). The dataset consists of 1868 FLS images captured using ARIS Explorer 3000 sensor. The objects used to produce this dataset contain typical house-hold marine debris and distractor marine objects (tires, hooks, valves,etc), divided in 11 classes plus a background class. Performance of state of the art semantic segmentation architectures with a variety of encoders have been analyzed on this dataset and presented as baseline results. Since the images are grayscale, no pre-trained weights have been used. Comparisons are made using Intersection over Union (IoU). The best performing model is Unet with ResNet34 backbone at 0.7481 mIoU.
Cite
Text
Singh and Valdenegro-Toro. "The Marine Debris Dataset for Forward-Looking Sonar Semantic Segmentation." IEEE/CVF International Conference on Computer Vision Workshops, 2021. doi:10.1109/ICCVW54120.2021.00417Markdown
[Singh and Valdenegro-Toro. "The Marine Debris Dataset for Forward-Looking Sonar Semantic Segmentation." IEEE/CVF International Conference on Computer Vision Workshops, 2021.](https://mlanthology.org/iccvw/2021/singh2021iccvw-marine/) doi:10.1109/ICCVW54120.2021.00417BibTeX
@inproceedings{singh2021iccvw-marine,
title = {{The Marine Debris Dataset for Forward-Looking Sonar Semantic Segmentation}},
author = {Singh, Deepak and Valdenegro-Toro, Matias},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2021},
pages = {3734-3742},
doi = {10.1109/ICCVW54120.2021.00417},
url = {https://mlanthology.org/iccvw/2021/singh2021iccvw-marine/}
}