Reliable Detection of Doppelgängers based on Deep Face Representations
This addresses a specific reliability issue in facial recognition systems for security applications, but is incremental as it builds on existing databases and morphing techniques.
The paper tackles the problem of doppelgängers causing false matches in facial recognition systems, finding they significantly increase false match rates, and proposes a detection method that achieves a 2.7% equal error rate in distinguishing them from mated comparisons.
Doppelgängers (or lookalikes) usually yield an increased probability of false matches in a facial recognition system, as opposed to random face image pairs selected for non-mated comparison trials. In this work, we assess the impact of doppelgängers on the HDA Doppelgänger and Disguised Faces in The Wild databases using a state-of-the-art face recognition system. It is found that doppelgänger image pairs yield very high similarity scores resulting in a significant increase of false match rates. Further, we propose a doppelgänger detection method which distinguishes doppelgängers from mated comparison trials by analysing differences in deep representations obtained from face image pairs. The proposed detection system employs a machine learning-based classifier, which is trained with generated doppelgänger image pairs utilising face morphing techniques. Experimental evaluations conducted on the HDA Doppelgänger and Look-Alike Face databases reveal a detection equal error rate of approximately 2.7% for the task of separating mated authentication attempts from doppelgängers.