NABirds Dataset: Download it now!

Try out a dataset for fine-grained recognition, featuring 400 species of North America’s birds. 

NABirds V1 is a collection of 48,000 annotated photographs of the 400 species of birds that are commonly observed in North America. More than 100 photographs are available for each species, including separate annotations for males, females and juveniles that comprise 700 visual categories. This dataset is to be used for fine-grained visual categorization experiments.

• More than 550 visual categories, organized taxonomically
• Photos curated in collaboration with domain experts
  

For more information contact:

Grant Van Horn (gvanhorn@cornell.edu) 

NABirds is a collaborative effort between:

The NABirds dataset has been made possible by a collaboration among the Cornell Lab of Ornithology, Cornell Tech (S. Belongie), Caltech (P. Perona, G. Van Horn, S. Branson), Brigham Young University (R. Farrell) and thousands of contributing photographers and visitors to the All About Birds website.