Attention Aware Wavelet-based Detection of Morphed Face Images

TL;DR

This paper introduces a wavelet-based deep neural network with attention mechanisms for detecting morphed face images, focusing on salient facial regions in spectral space to improve security in face recognition systems.

Contribution

It presents a novel end-to-end trainable attention-based DNN that operates in wavelet space, specifically targeting facial landmarks for improved morph detection accuracy.

Findings

Effective detection on multiple datasets (VISAPP17, LMA, MorGAN)

Attention maps highlight regions critical for distinguishing genuine and morphed images

Ablation study confirms the importance of attention mechanisms in morph detection

Abstract

Morphed images have exploited loopholes in the face recognition checkpoints, e.g., Credential Authentication Technology (CAT), used by Transportation Security Administration (TSA), which is a non-trivial security concern. To overcome the risks incurred due to morphed presentations, we propose a wavelet-based morph detection methodology which adopts an end-to-end trainable soft attention mechanism . Our attention-based deep neural network (DNN) focuses on the salient Regions of Interest (ROI) which have the most spatial support for morph detector decision function, i.e, morph class binary softmax output. A retrospective of morph synthesizing procedure aids us to speculate the ROI as regions around facial landmarks , particularly for the case of landmark-based morphing techniques. Moreover, our attention-based DNN is adapted to the wavelet space, where inputs of the network are coarse-to-fine spectral representations, 48 stacked wavelet sub-bands to be exact. We evaluate performance of the proposed framework using three datasets, VISAPP17, LMA, and MorGAN. In addition, as attention maps can be a robust indicator whether a probe image under investigation is genuine or counterfeit, we analyze the estimated attention maps for both a bona fide image and its corresponding morphed image. Finally, we present an ablation study on the efficacy of utilizing attention mechanism for the sake of morph detection.