Phase-aggregated Dual-branch Network for Efficient Fingerprint Dense Registration

TL;DR

This paper introduces PDRNet, a dual-branch deep learning model that improves fingerprint dense registration accuracy and robustness, especially for low-quality samples, while maintaining efficiency through multi-stage correlation and texture feature integration.

Contribution

The paper proposes a novel dual-branch network with multi-stage interactions for enhanced fingerprint registration, combining local detail perception with global stability.

Findings

Achieves state-of-the-art accuracy and robustness in fingerprint registration

Performs well on diverse and comprehensive fingerprint databases

Maintains competitive inference speed compared to existing methods

Abstract

Fingerprint dense registration aims to finely align fingerprint pairs at the pixel level, thereby reducing intra-class differences caused by distortion. Unfortunately, traditional methods exhibited subpar performance when dealing with low-quality fingerprints while suffering from slow inference speed. Although deep learning based approaches shows significant improvement in these aspects, their registration accuracy is still unsatisfactory. In this paper, we propose a Phase-aggregated Dual-branch Registration Network (PDRNet) to aggregate the advantages of both types of methods. A dual-branch structure with multi-stage interactions is introduced between correlation information at high resolution and texture feature at low resolution, to perceive local fine differences while ensuring global stability. Extensive experiments are conducted on more comprehensive databases compared to previous works. Experimental results demonstrate that our method reaches the state-of-the-art registration performance in terms of accuracy and robustness, while maintaining considerable competitiveness in efficiency.