skip to main content


Title: Identical Twins Face Morph Database Generation
By combining two or more face images of look-alikes, morphed face images are generated to fool Facial Recognition Systems (FRS) into falsely accepting multiple people, leading to failures in security systems. Despite several attempts in the literature, finding pairs of bona fide faces to generate the morphed images is still a challenging problem. In this paper, we morph identical twin pairs to generate extremely difficult morphs for FRS. We first explore three methods of morphed face generation, GAN-based, landmark-based, and a wavelet-based morphing approach. We leverage these methods to generate morphs from the identical twin pairs that retain high similarity to both subjects while resulting in minimal artifacts in the visual domain. To further improve the difficulty of recognizing morphed face images, we perform an ablation study to apply adversarial perturbation to the morphs such that they cannot be detected by trained morph classifiers. The evaluation of the generated identical twin-morphed dataset is performed in terms of vulnerability analysis and presentation attack error rates.  more » « less
Award ID(s):
1650474
NSF-PAR ID:
10401290
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
Identical Twins Face Morph Database Generation
Page Range / eLocation ID:
1 to 9
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Morph images threaten Facial Recognition Systems (FRS) by presenting as multiple individuals, allowing an adversary to swap identities with another subject. Morph generation using generative adversarial networks (GANs) results in high-quality morphs unaffected by the spatial artifacts caused by landmark-based methods, but there is an apparent loss in identity with standard GAN-based morphing methods. In this paper, we propose a novel StyleGAN morph generation technique by introducing a landmark enforcement method to resolve this issue. Considering this method, we aim to enforce the landmarks of the morphed image to represent the spatial average of the landmarks of the bona fide faces and subsequently the morph images to inherit the geometric identity of both bona fide faces. Exploration of the latent space of our model is conducted using Principal Component Analysis (PCA) to accentuate the effect of both the bona fide faces on the morphed latent representation and address the identity loss issue with latent domain averaging. Additionally, to improve high frequency reconstruction in the morphs, we study the train-ability of the noise input for the StyleGAN2 model. 
    more » « less
  2. A morph is an image of an ambiguous subject generated by combining multiple individuals. The morphed image can be submitted to a facial recognition system and erroneously verified with the contributing bad actors. When submitted as a passport image, a morphed face poses a national security threat because a passport can then be shared between the individuals. As morphed images become easier to generate, it is vital that the research community expands available datasets in order to contentiously improve current technology. Children are a challenging paradigm for facial recognition systems and morphing children takes advantage of this disparity. In this paper, we morph juvenile faces in order to create a unique, high-quality dataset to challenge FRS. To the best of our knowledge, this is the first study on the generation and evaluation of juvenile morphed faces. The evaluation of the generated morphed juvenile dataset is performed in terms of vulnerability analysis and presentation attack error rates. 
    more » « less
  3. Morph detection is of paramount significance when the integrity of Automatic Face Recognition (AFR) systems are concerned. Considering the risks incurred by morphing attacks, a robust automated morph detector is required which can distinguish authentic bona fide samples from altered morphed images. We leverage the wavelet sub-band decomposition of an input image, yielding the fine-grained spatial-frequency content of the input image. To enhance the detection of morphed images, our goal is to find the most discriminative information across frequency channels and spatial domain. To this end, we propose an end-to-end attention-based deep morph detector which assimilates the most discriminative wavelet sub-bands of a given image which are obtained by a group sparsity representation learning scheme. Specifically, our group sparsity-constrained Deep Neural Network (DNN) learns the most discriminative wavelet sub-bands (channels) of an input image while the attention mechanism captures the most discriminative spatial regions of input images for the downstream task of morph detection. To this end, we adopt three attention mechanisms to diversify our refined features for morph detection. As the first attention mechanism, we employ the Convolutional Block Attention Module (CBAM) which provides us with refined feature maps. As the second attention mechanism, compatibility scores across spatial locations and output of our DNN highlight the most discriminative regions, and lastly, the multiheaded self-attention augmented convolutions account for our third attention mechanism. We evaluate the efficiency of our proposed framework through extensive experiments using multiple morph datasets that are compiled using bona fide images available in the FERET, FRLL, FRGC, and WVU Twin datasets. Most importantly, our proposed methodology has resulted in a reduction in detection error rates when compared with state-of-the-art results. Finally, to further assess our multi-attentional morph detection, we delve into different combinations of attention mechanisms via a comprehensive ablation study. 
    more » « less
  4. null (Ed.)
    The rise of the multiracial population has been met with a growing body of research examining multiracial face perception. A common method for creating multiracial face stimuli in past research has been mathematically averaging two monoracial “parent” faces of different races to create computer-generated multiracial morphs, but conclusions from research using morphs will only be accurate to the extent that morphs yield perceptual decisions similar to those that would be made with real multiracial faces. The current studies compared race classifications of real and morphed multiracial face stimuli. We found that oval-masked morphed faces were classified as multiracial significantly more often than oval-masked real multiracial faces (Studies 1 and 2), but at comparable levels to unmasked real multiracial faces (Study 2). Study 3 examined factors that could explain differences in how morphs and real multiracial faces are categorized and pointed to the potential role that unusualness/distinctiveness might play. 
    more » « less
  5. Abstract

    We assessed the status of two New Zealand endemic morphodemes in the genus Sticta, currently treated as two separate taxa, Sticta filix and Sticta lacera. Both are green-algal lichens with a distinct stipe that grow in forested habitats and are suitable indicators of the indigenous vegetation health in forest ecosystems in New Zealand. They exhibit different morphologies and substrate ecologies: S. filix forms rather robust thalli, often on exposed trunks of phorophytes, with erect stems distinctly emerging from the substrate, whereas S. lacera is a more delicate lichen growing near the base of trees, usually among bryophyte mats or sheltered in the exposed portions of the phorophyte root-plate, with a prostrate, branched, stolon-like stem barely emerging from the substrate. Throughout their range, both taxa grow sympatrically and often in close proximity (syntopically). Despite the differences, ITS barcoding does not support the two morphodemes as separate species. In this study we assessed two possible explanations: (1) S. filix and S. lacera are discrete phenotypes of a single species, caused by developmental switching triggered by a discrete environmental variable, the propagules developing either on bare substrate or between bryophytes; and (2) the two morphodemes represent separate lineages, but ITS does not provide sufficient resolution to reflect this. We performed a quantitative analysis of morphological and ecological parameters, based on vouchered herbarium collections and field observations on iNaturalist NZ (https://inaturalist.nz), to assess the level of discreteness of the growth forms and to test for a correlation with the presence of a bryophyte mat. We further took advantage of an existing molecular data set from a target capture approach, comprised of 205 protein markers. This data set was used to establish a framework of percentage identities between pairs of the same and of different species among lobarioid Peltigeraceae and then to test whether the S. filix/lacera pairing fell closer to a within-species or a between-species pairing. The morphometric analysis of herbarium material resolved S. filix and S. lacera as two discrete morphs with little overlap, supported by numerous observations on iNaturalist NZ. However, whereas herbarium material suggested a significant association of the lacera morph with bryophyte mats, no such pattern was evident from field images on iNaturalist NZ, in which both morphs frequently associated with bryophyte mats. This highlights the limitations of herbarium material to correctly assess substrate ecology, whereas iNaturalist NZ postings had issues with correct identifications, given that especially S. lacera is easily confused with Pseudocyphellaria multifida. Based on the target capture data, the percentage identity of the S. filix/lacera pairing (99.43%) was significantly higher than that of all 12 between-species pairings (93.20–98.01%); it was at the same time lower than that of all within-species pairings (99.63–99.99%) but significantly so only in comparison with five out of the eight within-species pairings. The target capture approach is thus inconclusive, but the combination of all data suggests that S. filix and S. lacera are not discrete morphodemes of a single species but represent two separate lineages which emerged recently and hence cannot be resolved using the ITS barcoding marker or even a deeper phylogenomic approach based on protein-coding markers. We propose transplantation experiments and the application of RADseq to further assess this situation.

     
    more » « less