Face analysis has been used widely in many applications, like in mobile payment, online banking. However, face data can be vulnerable to digital and physical attacks. In this research, we attempt to explore the new generation of various face forgery, adversarial and presentation attacks, improve face data integrity, develop robust and reliable learning based face analysis approaches, establish resource efficient and explainable learning methods against general attacks for trustable face analysis. This research is expected to have a significant impact in the practical and social aspects due to the wide applicability of autonomous systems to our daily life.? This research includes both theoretical analysis and experimental validations using publicly available datasets with heterogeneous and fragmented data. Prototype systems will be developed with the collaboration of industry.?? The research is carried out in the Center for Machine Vision and Signal Analysis, University of Oulu.?

2022

2024