Professor wins IARPA subcontract to improve long-range, whole-body biometric identification
A West Virginia University (WVU) professor has earned a four-year, $750,000 subcontract to refine the collection of whole-body biometric data at a distance as part of an ongoing project by a research organization of the U.S. Intelligence Community.
Jeremy Dawson, an associate professor of the Lane Department of Computer Science and Electrical Engineering, will form a research group with undergraduate and graduate students enrolled in the university to collect whole-body imaging data from hundreds of voluntary participants to address the challenges of identifying people at extreme distances and angles. The dataset will then be used to train and test biometric algorithms.
“Face recognition performance is severely impacted by distance, the individual’s pose, the lack of adequate resolution; all those things weigh heavily on how well facial recognition systems have been able to correctly identify people,” Dawson comments. “The data that we collect will enable the development of new face recognition algorithms that are robust enough to handle those challenging conditions that are present in in real world operational scenarios.”
Dawson says the biometric dataset will be collected at a “relatively close distance” with cameras, and then collect images from longer distances and extreme angles while the participants are performing specific actions like walking or interacting with their phone. “Gathering enough data in situations that would make it harder to recognize people is our primary focus,” Dawson says.
The aim is to create a large-scale biometric database of images and videos of a person’s face, gait, body shape, and type to refine the identification accuracy of a biometric algorithm from low-quality images gained from a drone or security camera. Dawson adds that the face and whole-body imagery will help improve human recognition as a whole across age, gender, and ethnicity, thus making more equitable algorithms. WVU says the research will have potential in healthcare, law enforcement, and national security.
The subcontract is part of the Biometric Recognition and Identification at Altitude and Range (BRIAR) program organized by Intelligence Advanced Research Projects Activity (IARPA), a research and development arm of the U.S. Office of the Director of National Intelligence (ODNI). IARPA issued contracts to companies and universities in March as part of its effort to identify and recognize individuals from a distance and disturbances like atmospheric turbulence using drones.
WVU is a subcontractor to Systems and Technology Research (STR), which won the BRIAR contract in March. The university will provide the biometric data to STR and IARPA, which will be then used as a resource for the general research community in biometrics.
ODNI has taken efforts since 2019 to hone whole-body biometric recognition at a distance. In 2020, it published a request for information and released more details for what it was seeking, like the capability to identify people with biometrics from 300 meters being addressed by Dawson.
biometric identification | Biometric Recognition and Identification at Altitude and Range (BRIAR) | biometrics | biometrics research | facial recognition | funding | gait recognition | IARPA | long-distance | research and development | West Virginia