Facial recognition tools are expanding at a rate that may excite or worry you, but now a collaboration between computer scientists and biologists have taken them in a different direction entirely. They have created a tool to allow field researchers to recognize individual lemurs, and possibly many other species too.
Humans have evolved to be very good at telling other members of our own species apart, and we do pretty well with cats and dogs, but when we get to less familiar creatures, it gets hard. Jane Goodall once faced resistance from older scientists for naming the chimpanzees she was studying, but today this is an automatic feature of animal behavioral research, and everyone involved needs to be able to match the face to the name.
In fact, primate researcher Vanessa Woods’ memoir stemmed in part from her inability to tell the difference between the capuchin monkeys she was studying. As a paper in BMC Zoology notes: “To address many evolutionary questions, however, it is necessary to collect data on known individuals over long periods of time,” making it particularly challenging to study growing animals.
Co-lead author Dr Rachel Jacobs of George Washington University thought technology might help. Jacobs used 462 photographs her team had taken of 80 red-bellied lemurs (Eulemur rubriventer) from Ranomafana National Park, Madagascar.
“Our co-author Anil Jain and members from his laboratory were able to adapt a facial recognition system designed for human faces so that it recognizes individual lemurs based on their facial characteristics,” Jacobs said in a statement. “We were surprised with the high degree of accuracy that we achieved, which shows that facial recognition can be a useful tool for lemur identification.”
The program proved 98.7 percent accurate.
Can you identify the lemur that robbed you? Recognizing a lemur in different lights isn't easy, but LemurFaceID almost always manages it. Crouse et al/BMCZoology
Red-bellied lemurs are just one of the almost 100 species of lemurs living in Madagascar, but are much better studied than most. Building the system, known as LemurFaceID, was made easier through the fact that images existed of all the red-bellied lemurs in the area. Distinguishing unknown interlopers from the known faces may prove more of a challenge for the technology.
Red-bellied lemurs have distinctive white patches under their eyes that can be used for identification, but some other lemur species may prove more difficult to recognize. Nevertheless, the authors hope to apply the work widely to mammals with distinctive facial hair, including bears, raccoons, and sloths.
The use of facial recognition technology could prevent the need to capture and collar study animals. Even when these do not injure the individual (or result in a nasty bite to the researcher), they can change the very behavior that is being studied.
We just thought you might like another picture of red-bellied lemurs. Atosan/Shutterstock
1
