Reliable Detection of Doppelg\"angers based on Deep Face Representations

TL;DR

This paper investigates the challenge doppelg"angers pose to facial recognition systems, demonstrating their high similarity scores and false match rates, and introduces a machine learning method to reliably detect them using deep face representations.

Contribution

It presents a novel doppelg"anger detection approach based on deep face features and machine learning, significantly reducing false matches caused by lookalikes.

Findings

High false match rates for doppelg"angers in face recognition systems

Proposed detection method achieves approximately 2.7% EER in experiments

Effective separation of mated attempts from doppelg"angers using deep representations

Abstract

Doppelg\"angers (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 doppelg\"angers on the HDA Doppelg\"anger and Disguised Faces in The Wild databases using a state-of-the-art face recognition system. It is found that doppelg\"anger image pairs yield very high similarity scores resulting in a significant increase of false match rates. Further, we propose a doppelg\"anger detection method which distinguishes doppelg\"angers 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 doppelg\"anger image pairs utilising face morphing techniques. Experimental evaluations conducted on the HDA Doppelg\"anger and Look-Alike Face databases reveal a detection equal error rate of approximately 2.7% for the task of separating mated authentication attempts from doppelg\"angers.