Face recognition that could once only identify humans is evolving to work on bears and cows. Bear biologist Melanie Clapham, who studies bears in Canada, said it can be hard for people to tell bears apart as they track them over the years because their appearances change drastically, CNN reported. Tracking individual bears helps with research and conservation.
A few years ago, Clapham began to wonder if facial recognition software, typically used for humans, could help. Clapham decided to get together with two Silicon Valley-based tech workers, Ed Miller and Mary Nguyen, and they created BearID, which uses facial-recognition software to monitor grizzly bears, CNN reported.
The software uses AI technology to recognize 132 animals individually thus far. Clapham told CNN their technology is less invasive than putting a collar or RFID tag on an animal. So far, BearID has gathered 4,674 images of grizzly bears, with 80% of the images used for training the facial-recognition system and 20% for testing it. The software is currently 84 percent accurate, according to Clapham.
"It does way better than we do," Miller told CNN.
Joe Hogland, a Kansas cattle rancher, is hoping to do something similar on cattle ranches with an app called CattleTracs. The app will reportedly be able to have people snap pictures of cattle with GPS coordinates and it will be stored in the app’s database.
Photos taken of the animal from then on would be able to track it over time.
Beef cattle pass through many different people and places during their lives, Hogland told CNN.
"Being able to trace that diseased animal, find its source, quarantine it, do contact tracing — all the things we're talking about with coronavirus are things we can do with animals, too," he said.