Datasets:

vapaau
/

jambo

date stringclasses 4 values	agreement float64 0.33 1	majority_label stringclasses 3 values	Bio1 stringclasses 3 values	Bio2 stringclasses 3 values	Bio3 stringclasses 3 values	CV1 stringclasses 3 values	CV2 stringclasses 3 values	CV3 stringclasses 3 values
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	0.666667	stone	stone	bad	bad	stone	stone	stone
10-11	0.666667	bad	stone	bad	bad	stone	bad	bad
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	bad	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	0.666667	stone	stone	stone	bad	stone	bad	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	sand	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	sand	sand	sand	sand	sand	sand	sand
10-11	1	sand	sand	sand	sand	sand	sand	sand
10-11	1	sand	sand	sand	sand	sand	sand	sand
10-11	0.833333	sand	sand	sand	sand	sand	stone	sand
10-11	0.5	sand	sand	bad	sand	bad	sand	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.5	bad	sand	bad	sand	bad	sand	bad
10-11	0.5	bad	sand	bad	sand	bad	sand	bad
10-11	0.833333	sand	sand	sand	sand	bad	sand	sand
10-11	0.5	sand	sand	bad	sand	bad	bad	sand
10-11	0.666667	stone	sand	stone	sand	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	0.833333	sand	sand	bad	sand	sand	sand	sand
10-11	0.833333	sand	sand	sand	sand	stone	sand	sand
10-11	0.833333	sand	sand	stone	sand	sand	sand	sand
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.666667	stone	stone	stone	bad	stone	stone	bad
10-11	0.666667	bad	stone	bad	bad	stone	bad	bad
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.666667	bad	stone	bad	bad	stone	bad	bad
10-11	0.666667	stone	sand	stone	sand	stone	stone	stone
10-11	0.666667	bad	sand	bad	sand	bad	bad	bad
10-11	0.666667	stone	stone	stone	bad	stone	stone	bad
10-11	0.833333	bad	stone	bad	bad	bad	bad	bad
10-11	0.666667	bad	stone	bad	bad	bad	sand	bad
10-11	0.666667	stone	stone	bad	bad	stone	stone	stone
10-11	0.666667	stone	stone	bad	sand	stone	stone	stone
10-11	0.833333	stone	stone	bad	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	0.5	stone	bad	stone	sand	bad	stone	stone
10-11	0.666667	stone	sand	stone	bad	stone	stone	stone
10-11	0.833333	bad	bad	bad	bad	bad	sand	bad
10-11	0.666667	stone	stone	stone	bad	bad	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	0.666667	bad	sand	bad	bad	bad	sand	bad
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.5	stone	stone	stone	bad	stone	sand	bad
10-11	0.833333	bad	bad	bad	bad	sand	bad	bad
10-11	0.5	stone	stone	bad	bad	stone	stone	bad
10-11	0.666667	stone	stone	bad	bad	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	0.833333	stone	stone	stone	bad	stone	stone	stone
10-11	1	bad	bad	bad	bad	bad	bad	bad
10-11	0.666667	sand	sand	stone	sand	sand	stone	sand
10-11	0.5	stone	sand	stone	sand	sand	stone	stone
10-11	0.833333	sand	sand	bad	sand	sand	sand	sand
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.5	sand	sand	stone	sand	stone	stone	sand
10-11	0.833333	stone	sand	stone	stone	stone	stone	stone
10-11	0.666667	stone	sand	stone	sand	stone	stone	stone
10-11	0.666667	stone	sand	stone	stone	stone	sand	stone
10-11	0.833333	stone	sand	stone	stone	stone	stone	stone
10-11	0.666667	sand	sand	stone	sand	sand	sand	stone
10-11	0.666667	stone	sand	stone	stone	bad	stone	stone
10-11	0.5	bad	stone	bad	bad	bad	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	0.5	sand	sand	stone	sand	stone	sand	bad
10-11	1	stone	stone	stone	stone	stone	stone	stone
10-11	1	stone	stone	stone	stone	stone	stone	stone

End of preview. Expand in Dataset Viewer.

The JAMBO dataset contains 3290 underwater images of the seabed captured by an ROV in temperate waters in the Jammer Bay area off the North West coast of Jutland, Denmark. All the images have been annotated by six annotators to contain one of three classes: sand, stone, or bad. The three classes are defined as follows:

Sand habitats are characterized as primarily sand or muddy sand with less than 5% clay and less than 30% cover of stones/boulders, vegetation, and mussel bed.
Stone reef habitats are characterized by having more than 30% seabed cover of stones or boulders.
Bad is a class used to label images that cannot be confidently annotated as containing one of the aforementioned habitat types by the annotator due to poor image quality, turbidity, or similar.

Each of the six annotators have labelled all the images (that is, six individual annotations are provided for each image), which allows for analyzing how inter-annotator disagreement can affect the performance of machine learning models. Cross-validation splits and date-based splits are provided in the jambo_splits_public.csv file. Check out the starter notebook howto_jambo.ipynb to get started.

For more information about the dataset and baseline models, please see the paper to be presented the ECCV 2024 Computer Vision for Ecology (CV4E) Workshop: Underwater Uncertainty: A Multi-Annotator Image Dataset for Benthic Habitat Classification

Downloads last month: 108

Size of downloaded dataset files:

4.39 GB

Size of the auto-converted Parquet files:

4.39 GB

Number of rows:

3,290

Collection including vapaau/jambo

Underwater & fishy datasets

Collection

2 items • Updated Sep 17