Facial recognition is a biometric modality that has gained interest over the last years and with some performances that are increasing. In the specific context of Border Control, face recognition is even more popular as it is often the best way to authenticate a traveller against its travel document (passport). Indeed, as the picture stored into the passport chip is readable by all countries that comply with the ICAO standards, this data can be extracted and used to perform a 1:1 face recognition.
So if face recognition is a modality that is practical both from the governmental side and the citizen side (contact less, seen as not intrusive, well received by the general public) it is a technology that still need to improve. All development related to the face extraction and normalization or aiming at reducing all the “noises” (pose variation, ageing, eight gain) and environment conditions will greatly enhance the performance and pave the way for a more robust solution in the future. The work performed by BODEGA project is clearly showing what improvement can be set up in terms of supporting tools as well as on the way the information is presented.
In this regards, BODEGA project has developed some mock up aiming at supporting the decision making process of the border guard officer by allowing him/her to use additional tool in case of doubt.