Technology tracks faces in great-ape footage
New artificial intelligence (AI) software developed by scientists at the University of Oxford can accurately recognise and track the faces of individual chimps in the wild.
According to Science Advances, the application will significantly reduce the amount of time and resources researchers spend analysing video footage.
“By harnessing the power of machine learning to unlock large video archives, it makes it feasible to measure how the social interactions of a group change over several generations,” says Dan Schofield, a researcher at the University of Oxford.
Previous primate facial recognition software required researchers to spend hours preparing large datasets before the model could perform. This advanced software has automated the process, generating 10 million face images from raw footage of wild chimpanzees in Guinea, West Africa.
“Using recognition systems to monitor species in the wild is just the starting point,” says Schofield. “In the future, the model could recognise specific behaviours.”
The current software is available opensource and has the potential to be applied to other animals: “With many species and ecosystems under threat, being able to automatically monitor the health and size of populations in the wild will be crucial for helping to allocate resources on the ground,” says Schofield. Harry Gray
FIND OUT MORE Science Advances: bit.ly/2m1x0wv