Idiap expanding its ‘Deepfake detection and attribution’ research team
The Idiap Research Institute has posted an announcement on their site looking for qualified candidates for a Ph.D. position researching ‘Deepfake detection and attribution.’
The project will be developed in Dr. Sebastien Marcel’s Biometric Security and Privacy lab at Idiap, and will focus on one class modeling, spatio-temporal learning, few-shot learning, and adversarial training.
Deepfake detection is an evolving field, with various technologies being used to assess the validity of video feeds.
Biometric techniques have been used by Binghamton University researchers, for example, to uncover deepfakes and determine the generators that created them.
Building on Bob, a free signal-processing and machine learning toolbox developed by the Biometrics group, the new team will now develop new solutions under the supervision of Prof. Christophe Champod from the University of Lausanne (UNIL). Involvement in open source libraries is therefore sought.
“The ideal candidate should hold a Master’s degree in Computer Science, Forensic Science or related fields,” the ad reads.
A background in statistics, applied mathematics, optimization, linear algebra, and signal processing will be needed for the position, as well as programming, scripting, and machine learning skills.
Upon successful selection of the candidate, the individual will be registered as a doctoral student at UNIL and will be working on the development of deepfake detection and attribution for four years, subject to a successful candidacy exam in the first year. The research project is expected to begin early next year.
The Idiap Research Institute also partnered with Global ID earlier this year to develop contactless vein biometrics for hospital deployment, in an example of the Institute’s wide-ranging biometrics work.
For more information about the position and to apply for it, you can follow this link here.