The DLR-ACD dataset is a collection of aerial images for crowd counting and density estimation, as well as for person localization at mass events. It contains 33 large aerial images acquired through 16 different flight campaigns at various mass events and over urban scenes involving crowds, such as sport events, city centers, open-air fairs and festivals.
The images were captured with standard DSLR cameras installed on a helicopter, and their spatial resolution (or ground sampling distance – GSD) ranges from 4.5 to 15 cm/pixel. The dataset was labeled manually with point-annotations on individual people and contains 226,291 person annotations in total, ranging from 285 to 24,368 annotations per image. DLR-ACD has been used and published in . Please cite  if you use it in your work.
 R. Bahmanyar, E. Vig, and P. Reinartz, "MRCNet: Crowd Counting and Density Map Estimation in Aerial and Ground Imagery," in BMVC Workshop on Object Detection and Recognition for Security Screening, September 2019.