The European Biometrics Research Award goes to Tiago de Freitas Pereira, PostDoc researcher at Idiap

Face recognition systems have been proposed over the last three decades. Ranging from visual light images to infrared ones, various technologies are available. Being able to mix various recognition methods is a key element to provide a more accurate solution. In his thesis Tiago de Freitas Pereira from Idiap’s Biometrics security & privacy group looked how to combine these various technologies to apply them in so called heterogeneous face recognition situations. The key difficulty in the comparison of faces in such conditions is that images from the same subject may differ in appearance due to changes in image domain. Tiago’s work proposes a method to address this difficulty.

Using mathematical tools

In his research, Tiago analysed if the features used in visual light face recognition situation can also be applied in other light situations. Then using various mathematical tools, he showed that data from these facial recognition systems can be compared and used together for facial recognition. Then, large-scale experiments were conducted on several heterogeneous face recognition databases, covering various image domains showing competitive performances.

Moreover, the implementation of all the proposed techniques is integrated into a collaborative open source software library called Bob. The aim is to enforces fair evaluations and encourages reproducible research.

Link to the publication: Learning How To Recognize Faces in Heterogeneous Environments

European Association for Biometrics: https://www.eab.org/

EAB award Hall of Fame: https://www.eab.org/award/hall_of_fame.htm