Jungle primate footage once unusable now becomes research data thanks to an AI system
AI system tracks jungle primates through overgrown vegetation & poor lighting conditions that have foiled researchers using older methods
Primates in their natural habitats are getting new opportunities for behavioral studies from hours of video footage that previously were too degraded for human analysis with a computer vision tool called PriMAT as reported by wildlife coverage recently.
The PriMAT uses what researchers call "dynamic bounding boxes" to keep track of specific primates even when they move through densely planted areas and varying light patterns that typically make hours of recorded material useless for behavioral studies due to obstruction from plant life.
Get the latest model rankings, product launches, and evaluation insights delivered to your inbox.
This new method is changing how field researchers collect data. Where previous methods would discard blurry or obstructed sections of footage; researchers can now process footage that was rejected based on quality issues under prior video analysis systems.
PriMAT allows scientists to identify each animal within groups which is a game-changer for behavioral studies because it enables researchers to create detailed activity profiles of animals from continuous footage instead of relying on periodic clear observations.
Development comes at a time when there are growing pressures on wildlife researchers to study populations of primates affected by habitat loss and warming climates. Manual observation requires extensive human hours to review footage frame by frame (a process that makes no sense with poor-quality video).
PriMAT promises to address an obvious bottleneck in primate behavioral studies where field conditions have always limited the amount of usable data researchers could extract from recording efforts.