iWarpGAN: Disentangling Identity and Style to Generate Synthetic Iris Images

View on arXiv ← Back to list

Authors: Shivangi Yadav, Arun Ross

Published: 2023-05-21 23:10:14+00:00

AI Summary

iWarpGAN is a novel GAN-based method for generating synthetic iris images that disentangles identity and style, addressing limitations of existing methods in generating diverse and high-quality iris deepfakes. It uses two transformation pathways (identity and style) to generate iris images with both inter- and intra-class variations, improving the performance of deep learning-based iris matchers when used for data augmentation.

Abstract

Generative Adversarial Networks (GANs) have shown success in approximating complex distributions for synthetic image generation. However, current GAN-based methods for generating biometric images, such as iris, have certain limitations: (a) the synthetic images often closely resemble images in the training dataset; (b) the generated images lack diversity in terms of the number of unique identities represented in them; and (c) it is difficult to generate multiple images pertaining to the same identity. To overcome these issues, we propose iWarpGAN that disentangles identity and style in the context of the iris modality by using two transformation pathways: Identity Transformation Pathway to generate unique identities from the training set, and Style Transformation Pathway to extract the style code from a reference image and output an iris image using this style. By concatenating the transformed identity code and reference style code, iWarpGAN generates iris images with both inter- and intra-class variations. The efficacy of the proposed method in generating such iris DeepFakes is evaluated both qualitatively and quantitatively using ISO/IEC 29794-6 Standard Quality Metrics and the VeriEye iris matcher. Further, the utility of the synthetically generated images is demonstrated by improving the performance of deep learning based iris matchers that augment synthetic data with real data during the training process.

Key findings

iWarpGAN generates higher-quality synthetic iris images compared to existing GAN methods, as measured by ISO/IEC 29794-6 Standard Quality Metrics and VeriEye rejection rates. Augmenting real iris datasets with synthetic data generated by iWarpGAN improves the performance of deep learning-based iris matchers.

Approach

iWarpGAN uses two transformation pathways: an Identity Transformation Pathway to generate unique identities and a Style Transformation Pathway to control the style of the generated iris image. It concatenates these transformed codes to generate iris images with both inter- and intra-class variations. The model is trained using a combination of adversarial loss and reconstruction losses to ensure image quality and identity uniqueness.

Datasets

CASIA-Iris-Thousand, CASIA Cross Sensor Iris Dataset (CSIR), IITD-iris

Model(s)

Generative Adversarial Network (GAN) with RBF-based warp function for identity transformation and a style encoder for style transformation. Compared against WGAN, RaSGAN, and CITGAN.

Author countries

USA

← Previous