This dataset contains the data necessary to recreate the counting procedure described in the manuscript "Combining artificial intelligence and local ecological knowledge to document the largest ever-counted Cape buffalo mega-herd". The folder mababe-megaherd-data includes all data necessary for the automated detection, tracking and counting procedure, count forecasting and accuracy assessments, as well as the files generated during this procedure, including the supplementary videos referred to in the manuscript. Code for implementing the procedure are provided in the corresponding GitHub repository.