Research Areas

Face Recognition

As a popular biometrics modality, research on face recognition focuses on recognising faces subject to wild, uncontrolled environmental imaging conditions or changes in sensory hardware configurations in a one-to-one (verification/pair-matching) or one-to-many (identification) matching scenario.

Face Presentation Attack Detection

Research on face presentation attack detection is directed towards developing novel methodologies to secure face authentication systems against security threats and attacks made by presenting fake biometrics traits against system sensor to gain illegal access to the system.

Face Manipulation Detection

Face manipulation (an instance of digital image tampering) detection is an important topic in digital media forensics which focuses on assessing the authenticity of face images/videos with various potential applications such as assessing the credibility of news, digital communications, law enforcement, etc.

Novelty Detection

Novelty detection deals with the problem of identifying patterns which conform to a specific behaviour and distinguishing them from all other objects, typically known as novelties, outliers or anomalies. It serves as the core of a wide variety of applications such as intrusion detection, fraud detection, insurance, health care, surveillance, etc.

A summary of my research on face image analysis may be found in these slides.