Generative Iris Prior Embedded Transformer for Iris Restoration

TL;DR

This paper introduces Gformer, a novel Transformer-based model with a generative iris prior for restoring degraded iris images, significantly enhancing iris recognition accuracy.

Contribution

It presents a hierarchical encoder-decoder Transformer model that integrates a pretrained iris GAN prior for improved iris restoration.

Findings

Gformer outperforms existing methods in iris restoration.

Restored iris images lead to higher recognition accuracy.

The model effectively captures long-range dependencies in iris images.

Abstract

Iris restoration from complexly degraded iris images, aiming to improve iris recognition performance, is a challenging problem. Due to the complex degradation, directly training a convolutional neural network (CNN) without prior cannot yield satisfactory results. In this work, we propose a generative iris prior embedded Transformer model (Gformer), in which we build a hierarchical encoder-decoder network employing Transformer block and generative iris prior. First, we tame Transformer blocks to model long-range dependencies in target images. Second, we pretrain an iris generative adversarial network (GAN) to obtain the rich iris prior, and incorporate it into the iris restoration process with our iris feature modulator. Our experiments demonstrate that the proposed Gformer outperforms state-of-the-art methods. Besides, iris recognition performance has been significantly improved after applying Gformer.