An official website of the United States government
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
Landmark Enforcement and Style Manipulation for Generative Morphing
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
- Award ID(s):
- 1650474
- PAR ID:
- 10401291
- Date Published:
- Journal Name:
- 2022 IEEE International Joint Conference on Biometrics (IJCB), Abu Dhabi, United Arab Emirates, 2022
- Page Range / eLocation ID:
- 1 to 10
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
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
-
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
-
This paper proposes a framework for a privacy-safe iris presentation attack detection (PAD) method, designed solely with synthetically-generated, identity-leakage-free iris images. Once trained, the method is evaluated in a classical way using state-of-the-art iris PAD benchmarks. We designed two generative models for the synthesis of ISO/IEC 19794-6-compliant iris images. The first model synthesizes bona fide-looking samples. To avoid "identity leakage," the generated samples that accidentally matched those used in the model’s training were excluded. The second model synthesizes images of irises with textured contact lenses and is conditioned by a given contact lens brand to have better control over textured contact lens appearance when forming the training set. Our experiments demonstrate that models trained solely on synthetic data achieve a lower but still reasonable performance when compared to solutions trained with iris images collected from human subjects. This is the first-of-its-kind attempt to use solely synthetic data to train a fully-functional iris PAD solution, and despite the performance gap between regular and the proposed methods, this study demonstrates that with the increasing fidelity of generative models, creating such privacy-safe iris PAD methods may be possible. The source codes and generative models trained for this work are offered along with the paper.more » « less
-
We propose a novel unsupervised generative model that learns to disentangle object identity from other low-level aspects in class-imbalanced data. We first investigate the issues surrounding the assumptions about uniformity made by InfoGAN [10], and demonstrate its ineffectiveness to properly disentangle object identity in imbalanced data. Our key idea is to make the discovery of the discrete latent factor of variation invariant to identity-preserving transformations in real images, and use that as a signal to learn the appropriate latent distribution representing object identity. Experiments on both artificial (MNIST, 3D cars, 3D chairs, ShapeNet) and real-world (YouTube-Faces) imbalanced datasets demonstrate the effectiveness of our method in disentangling object identity as a latent factor of variation.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/IJCB54206.2022.10008001
