Iris-GAN: Learning to Generate Realistic Iris Images Using Convolutional GAN

TL;DR

This paper introduces Iris-GAN, a convolutional GAN framework that generates highly realistic and diverse iris images, outperforming classical models in capturing complex textures from training datasets.

Contribution

The paper presents a novel GAN-based approach specifically designed for realistic iris image synthesis, demonstrating improved realism and diversity over traditional statistical models.

Findings

Generated iris images are highly realistic and diverse.

The method captures complex iris textures effectively.

Images resemble the distribution of training datasets.

Abstract

Generating iris images which look realistic is both an interesting and challenging problem. Most of the classical statistical models are not powerful enough to capture the complicated texture representation in iris images, and therefore fail to generate iris images which look realistic. In this work, we present a machine learning framework based on generative adversarial network (GAN), which is able to generate iris images sampled from a prior distribution (learned from a set of training images). We apply this framework to two popular iris databases, and generate images which look very realistic, and similar to the image distribution in those databases. Through experimental results, we show that the generated iris images have a good diversity, and are able to capture different part of the prior distribution.