Snapshot Serengeti: high-frequency annotated camera trap images of 40 mammalian species

Swanson and colleagues published the Snapshot Serengeti dataset, one of the first large-scale, openly available, expertly verified camera-trap image collections, comprising millions of images from a dense camera grid in Serengeti National Park labeled by tens of thousands of citizen-science volunteers. Beyond its ecological value for studying the savanna mammal community, the dataset became the standard benchmark for training and testing machine-learning models on camera-trap imagery. Its careful documentation of volunteer consensus and species annotations enabled reproducible computer-vision research. The dataset directly enabled landmark deep-learning studies and remains a touchstone resource. It exemplified the value of open data and crowdsourcing in ecology and is heavily cited across both wildlife and artificial-intelligence literatures.