# Minutiae-Guided Fingerprint Embeddings via Vision Transformers

arXiv: https://arxiv.org/abs/2210.13994 · HTML: https://mlsanity.com/paper/2210.13994/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 31 references that are in the map.
- First posted on arXiv 25 Oct 2022.

## Builds on (most cited first, 13 of 13)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [Learning Face Representation from Scratch](https://mlsanity.com/paper/1411.7923/) (2014)
- [General Facial Representation Learning in a Visual-Linguistic Manner](https://mlsanity.com/paper/2112.03109/) (2021)
- [TransFER: Learning Relation-aware Facial Expression Representations with Transformers](https://mlsanity.com/paper/2108.11116/) (2021)
- [Transformer-based Multimodal Information Fusion for Facial Expression Analysis](https://mlsanity.com/paper/2203.12367/) (2022)
- [Face Transformer for Recognition](https://mlsanity.com/paper/2103.14803/) (2021)
- [PrintsGAN: Synthetic Fingerprint Generator](https://mlsanity.com/paper/2201.03674/) (2022)
- [Secure Face Matching Using Fully Homomorphic Encryption](https://mlsanity.com/paper/1805.00577/) (2018)
- [HERS: Homomorphically Encrypted Representation Search](https://mlsanity.com/paper/2003.12197/) (2020)
- [Fingerprint Feature Extraction by Combining Texture, Minutiae, and Frequency Spectrum Using Multi-Task CNN](https://mlsanity.com/paper/2008.11917/) (2020)
- [Transformer based Fingerprint Feature Extraction](https://mlsanity.com/paper/2209.03846/) (2022)

## Built on by (most cited first, 9 of 9)

- [AFR-Net: Attention-Driven Fingerprint Recognition Network](https://mlsanity.com/paper/2211.13897/) (2022)
- [IFViT: Interpretable Fixed-Length Representation for Fingerprint Matching via Vision Transformer](https://mlsanity.com/paper/2404.08237/) (2024)
- [Benchmarking fixed-length Fingerprint Representations across different Embedding Sizes and Sensor Types](https://mlsanity.com/paper/2307.08615/) (2023)
- [Mobile Contactless Palmprint Recognition: Use of Multiscale, Multimodel Embeddings](https://mlsanity.com/paper/2401.08111/) (2024)
- [Latent Fingerprint Matching via Dense Minutia Descriptor](https://mlsanity.com/paper/2405.01199/) (2024)
- [MRA-GNN: Minutiae Relation-Aware Model over Graph Neural Network for Fingerprint Embedding](https://mlsanity.com/paper/2307.16416/) (2023)
- [ViT Unified: Joint Fingerprint Recognition and Presentation Attack Detection](https://mlsanity.com/paper/2305.07602/) (2023)
- [Ridgeformer: Mutli-Stage Contrastive Training For Fine-grained Cross-Domain Fingerprint Recognition](https://mlsanity.com/paper/2506.01806/) (2025)
- [LatentPrintFormer: A Hybrid CNN-Transformer with Spatial Attention for Latent Fingerprint identification](https://mlsanity.com/paper/2511.08119/) (2025)

## Abstract (arXiv)

Minutiae matching has long dominated the field of fingerprint recognition. However, deep networks can be used to extract fixed-length embeddings from fingerprints. To date, the few studies that have explored the use of CNN architectures to extract such embeddings have shown extreme promise. Inspired by these early works, we propose the first use of a Vision Transformer (ViT) to learn a discriminative fixed-length fingerprint embedding. We further demonstrate that by guiding the ViT to focus in on local, minutiae related features, we can boost the recognition performance. Finally, we show that by fusing embeddings learned by CNNs and ViTs we can reach near parity with a commercial state-of-the-art (SOTA) matcher. In particular, we obtain a TAR=94.23% @ FAR=0.1% on the NIST SD 302 public-domain dataset, compared to a SOTA commercial matcher which obtains TAR=96.71% @ FAR=0.1%. Additionally, our fixed-length embeddings can be matched orders of magnitude faster than the commercial system (2.5 million matches/second compared to 50K matches/second). We make our code and models publicly available to encourage further research on this topic: https://github.com/tba.