In a breakthrough for computer vision and for bird watching, researchers and bird enthusiasts have enabled computers to achieve a task that stumps most humans—identifying hundreds of bird species pictured in photos. Build in concert with the exceedingly popular Merlin app, the Merlin Bird ID tool lets you upload an image of a bird that you’ve photographed, and if the photo shows one of the supported species, it returns the correct species in the top 3 results, 90% of the time. It currently supports 400 species in North America, but will eventually be expanding to more species in North America, and worldwide.
To see if Merlin can identify the bird in your photo, you upload an image and tell Merlin where and when you took it. To orient Merlin, you draw a box around the bird and click on its bill, eye, and tail. Merlin does the rest.
It compares characteristics of the bird to the expected species for the location based on eBird data, thanks to eBirders like you. By checking the appearance of the bird against eBird sightings, Merlin is able to return results that factor in occurrence patterns as well as the way the bird looks – a new paradigm in automated identification.
For this beta-testing phase, Photo ID is just available on the Web and does not work on phones or tablets—please bear with us. Merlin Bird Photo ID works through machine learning techniques, so it gets “smarter” the more people use it—please give it a try and help it improve!
Merlin’s computer vision system was created by the Visipedia research team at Cornell Tech and Caltech, in collaboration with the Cornell Lab of Ornithology. Their work was made possible with generous support from the Jacobs Technion-Cornell Institute at Cornell Tech, Google, and the National Science Foundation.