MorDeephy: Face Morphing Detection Via Fused Classification

• DOI: 10.48550/arxiv.2208.03110
• Publication Date: 2022-01-01
• Authors: Iurii Medvedev; Farhad Shadmand; Nuno Gonçalves
• Keywords: Morphing; Benchmark (surveying)

Face morphing attack detection (MAD) is one of the most challenging tasks in field face recognition nowadays. In this work, we introduce a novel deep learning strategy for single image detection, which implies discrimination morphed images along with sophisticated task complex classification scheme. It directed onto facial features, carry information about authenticity these features. Our work also introduces several additional contributions: public and easy-to-use benchmark results our wild datasets filtering strategy. method, call MorDeephy, achieved state art performance demonstrated prominent ability generalising to unseen scenarios.