The University of Southern California’s Information Sciences Institute has launched a government-funded study aimed at safeguarding protect biometrics systems from spoofing attacks, USC reported Thursday.
Institute researchers from USC’s Visual Intelligence and Multimedia Analytics group bring computer vision, voice recognition and other artificial intelligence capabilities to support spoofing prevention under the Biometric Authentication with Timeless Learner project.
The Intelligence Advanced Research Projects Activity’s Odin program sponsors the BATL project as part of an effort to invest in cutting-edge intelligence research.
Wael AbdAlmageed, a professor at USC’s Viterbi School of Engineering and research director at ISI, leads the VIMAL team for the collaborative effort with Swiss nonprofit Idiap Research Institute.
USC has achieved research milestones in the biometrics field since the creation of the project’s proposal five years ago.
Among these milestones is VIMAL’s development of an algorithm that helps users understand decisions generated by the biometric anti-spoofing system through natural language processing.
The Institute of Electrical and Electronics Engineers’ 2021 International Conference on Automatic Face and Gesture Recognition will feature a presentation of VIMAL’s research.