Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper analyzes differential morphing attack detection (D-MAD) and face recognition (FR), arguing they perform similar tasks. The authors demonstrate that current decision thresholds in FR systems make them vulnerable to morphing attacks, explaining the performance tradeoff between normal images and attack vulnerability. They propose leveraging existing FR systems for morphing detection using a novel evaluation threshold that limits vulnerability to morphing attacks, even of unknown types.
Face recognition systems are inherently vulnerable to morphing attacks due to suboptimal decision thresholds, but a simple threshold adjustment can fix it.
Morphing is a challenge to face recognition (FR) for which several morphing attack detection solutions have been proposed. We argue that face recognition and differential morphing attack detection (D-MAD) in principle perform very similar tasks, which we support by comparing an FR system with two existing D-MAD approaches. We also show that currently used decision thresholds inherently lead to FR systems being vulnerable to morphing attacks and that this explains the tradeoff between performance on normal images and vulnerability to morphing attacks. We propose using FR systems that are already in place for morphing detection and introduce a new evaluation threshold that guarantees an upper limit to the vulnerability to morphing attacks - even of unknown types.