Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancement-Driven Pretraining for Robust Fingerprint Representation Learning

Ekta Gavas, Kaustubh Olpadkar, Anoop Namboodiri

TL;DR

The paper tackles robust fingerprint verification under degraded conditions by introducing enhancement-driven, self-supervised pretraining using a U-Net encoder to learn discriminative fingerprint representations. It then evaluates these representations via a frozen encoder with a small MLP in a Siamese-style verification setup, showing improvements over established SSL baselines on public datasets. A key finding is that meaningful fingerprint features can be learned from degraded images without relying on enhanced samples, suggesting practical benefit for real-world biometric systems. The work lays groundwork for further improvements through additional real-world datasets and refined probing strategies to enhance generalizability.

Abstract

Fingerprint recognition stands as a pivotal component of biometric technology, with diverse applications from identity verification to advanced search tools. In this paper, we propose a unique method for deriving robust fingerprint representations by leveraging enhancement-based pre-training. Building on the achievements of U-Net-based fingerprint enhancement, our method employs a specialized encoder to derive representations from fingerprint images in a self-supervised manner. We further refine these representations, aiming to enhance the verification capabilities. Our experimental results, tested on publicly available fingerprint datasets, reveal a marked improvement in verification performance against established self-supervised training techniques. Our findings not only highlight the effectiveness of our method but also pave the way for potential advancements. Crucially, our research indicates that it is feasible to extract meaningful fingerprint representations from degraded images without relying on enhanced samples.

Enhancement-Driven Pretraining for Robust Fingerprint Representation Learning

TL;DR

The paper tackles robust fingerprint verification under degraded conditions by introducing enhancement-driven, self-supervised pretraining using a U-Net encoder to learn discriminative fingerprint representations. It then evaluates these representations via a frozen encoder with a small MLP in a Siamese-style verification setup, showing improvements over established SSL baselines on public datasets. A key finding is that meaningful fingerprint features can be learned from degraded images without relying on enhanced samples, suggesting practical benefit for real-world biometric systems. The work lays groundwork for further improvements through additional real-world datasets and refined probing strategies to enhance generalizability.

Abstract

Fingerprint recognition stands as a pivotal component of biometric technology, with diverse applications from identity verification to advanced search tools. In this paper, we propose a unique method for deriving robust fingerprint representations by leveraging enhancement-based pre-training. Building on the achievements of U-Net-based fingerprint enhancement, our method employs a specialized encoder to derive representations from fingerprint images in a self-supervised manner. We further refine these representations, aiming to enhance the verification capabilities. Our experimental results, tested on publicly available fingerprint datasets, reveal a marked improvement in verification performance against established self-supervised training techniques. Our findings not only highlight the effectiveness of our method but also pave the way for potential advancements. Crucially, our research indicates that it is feasible to extract meaningful fingerprint representations from degraded images without relying on enhanced samples.
Paper Structure (14 sections, 4 figures, 5 tables)

This paper contains 14 sections, 4 figures, 5 tables.

Table of Contents

  1. INTRODUCTION
  2. Contributions
  3. Related Work
  4. Image Enhancement
  5. Self-supervised Learning Techniques
  6. METHODOLOGY
  7. U-Net-based Pretraining
  8. Learning Fingerprint Representation
  9. EXPERIMENTS
  10. Datasets and Preprocessing
  11. Implementation Details
  12. Results
  13. LIMITATIONS AND FUTURE WORK
  14. CONCLUSION

Figures (4)

  • Figure 1: a) Architecture with verification objective i.e with binary classifier (at training and inference) b) Architecture to compute similarity scores (at inference). The dotted arrows indicate networks having tied weights (siamese network structure).
  • Figure 2: U-Net architecture for enhancement task for the pre-training stage in the self-supervised setting. For representation learning, the decoder is discarded and the binary classifier is attached.
  • Figure 3: Degraded (top row) and Enhanced (bottom row) image pairs on FVC dataset from enhancement pre-training
  • Figure 4: ROC curve based on similarity scores on SFinGe dataset(left) and FVC dataset (right)