flatbug-dataset a compilation of dataset of terrestrial arthropodes on various surfaces

This dataset is a compilation of 23 datasets all individually described in their respective DOI entry. The complete dataset was used to train "flatbug" a generalist instance segmentation algorithm that detects insects on a wide range of flat surfaces, and image dimensions and performs well in crowded cases.

DOIs to each dataset is in the "metadata.csv" file (also rendered below). There is a total of  6028 images. Each dataset is stored in its own subdirectory (with the same name) and contain its own COCO file (.json) that defines instances.