A Deep-Learning Approach for Visual Detection of an AUV Docking Station - Dataset

This dataset was used to train the models from the paper "A Deep-Learning Approach for Visual Detection of an AUV Docking Station" published by Ahmad et al. at Oceans 2024 Conference in Halifax.

# Dataset

The dataset contains:

1. images from Abisko Lake in Sweden[1]

2. images recorded in the Maritime Hall basin at DFKI

# File contents

Each of these datasets are put into seperate directories. The images were annotated using CVAT[2].

The dataset has been exported into the following formats:

1. YOLO

2. PascalVOC

3. COCO

The exported datasets does not contain raw images, rather they are places into a seperate zip folder.

# References

[1]: https://zenodo.org/record/7035132#.ZDfKE5FBzJU

[2]: https://www.cvat.ai/