Biometrics are techniques which can automatically establish a person’s identity through physiological measurements of their voice, face, iris and fingerprints along with behavioural characteristics such as gait, signature and keystrokes. They have a wide range of applications. They can be used to control physical access to buildings and other facilities, or to control virtual access to computers, mobile phones and bank accounts, or to personalise audiovisual content. In forensic applications, they can also be leveraged to gather fingerprints or DNA samples from crime scenes. However, they pose risks as they make it possible to harvest a large amount of data on individuals which can lead to greater control of populations by governments and businesses. Proper regulation of these technologies is therefore crucial.
These problems will be discussed in class along with technical content on the architecture, algorithms and evaluation of the most commonly used biometrics. The session will conclude with a short practical experiment.
Addressed to:
-People interested in artificial intelligence applications. Basic knowledge of machine/deep learning is preferable but not necessary.
-Business and systems managers and analysts.
