TACO is an open image dataset of waste in the wild. It contains photos of litter taken under diverse environments, from tropical beaches to London streets. These images are manually labeled and segmented according to a hierarchical taxonomy to train and evaluate object detection algorithms.
The dataset currently has 715 images with 2152 annotations and 3167 new images that need to be annotated. These annotations are labeled in 60 categories which belong to 28 super (top) categories. The data is annotated in the COCO format.
No assiciated paper or article has been found in regard to this dataset.
The data has been annotated using bounding boxes, object segmentation, background annotation and object context tag.
You can find the COCO annotation template here.
For more information regarding the classes used in annotating the data we suggest consulting the annotation stats and taxonomy webpages.
Sample annotated image:
Information regarding download can be found on the dataset's official github repo.
No associated model has been provided.
No benchmarks have been provided.
No associated challenges found in regard to this dataset.
The dataset is licenced under the CC By 4.0 licence.