Spoofing – The widespread use of personal verification systems based on fingerprints has shown some weaknesses related to the problem of security. Among the others, it is well-known that a fingerprint verification system can be deceived by submitting artificial reproductions of fingerprints made up of silicon or gelatine to the electronic capture device (optical, capacitive, etc…). These images are then processed as “true” fingerprints.

Liveness Detection – Therefore, a recent issue in the field of security in fingerprint verification (unsupervised especially) is known as “liveness detection” or “presentation attack detection” (PAD). The standard verification system is coupled with additional hardware or software modules aimed to certify the authenticity of the submitted fingerprints. Whilst hardware-based solutions are the most expensive, software-based ones attempt to measure liveness from characteristics of images themselves by simply applying image processing algorithms.

Software Liveness Classification – The problem of vitality detection is treated as a two-class classification problem (live or fake). An appropriate classifier is designed in order to extract the probability of the image vitality given the extracted set of features. LivDet2025 competition is open to all academic and industrial institutions which have a solution for software-based fingerprint recognition and liveness detection. 

Contact and contactless solutions – As the world moves towards more versatile and user-friendly security solutions, there is a great demand for advancements in both contact and contactless fingerprint sensors. Recognising this trend, LivDet2025 is committed to fostering innovation that bridges the gap between traditional contact-based systems and the new field of contactless fingerprint technology. This competition will focus on finger captured from fingerprint sensors and smartphones.

LivDet 2025 Competition Overview – This edition of LivDet 2025 has three challenges, applicable to both contact and contactless solutions

• Challenge 1. Liveness Detection in Action⁽¹⁾: Fingerprint Liveness Detection systems are not designed to operate stand-alone, but as a part of a recognition system. Competitors are invited to submit a complete algorithm able not only to output the probability of the image vitality (the so-called “liveness score”) given the extracted set of features but also an integrated match score (“integrated score”) which includes the probability above with the probability of belonging to the declared user. For this challenge, participants can decide whether to exploit the additional information coming from the enrolled user (“user-specific effect”⁽²⁾). During the design phase of the integrated system, they have at their disposal the new tool “Bio-WISE: Biometric recognition with integrated pad: simulation environment”. BIO-WISE is born to test the performance of fingerprint PAD and matchers combined sequentially⁽³⁾.
• Challenge 2. Fingerprint representation: In modern biometric systems, the compactness and the discriminability of feature vectors are fundamental to guarantee high performance in terms of accuracy and speed. Competitors are invited to submit a liveness detection algorithm which returns in addition to the probability of liveness, the feature vector corresponding to the input image with a maximum size of 512 bytes. The algorithms will be assessed on the basis of system accuracy and speed on PC with specified characteristics (see the page related to the challenge 2).
• Challenge 3. Adversarial Robustness: An insidious flip side of ML-based PAD solutions, is represented by adversarial attacks, namely, procedures intended to mislead a target detector. Recent works highlighted the possibility of transferring a fingerprint adversarial attack from the digital domain to the physical one: this produces new PAs that have a higher chance of passing PAD control. Competitors are challenged to develop a PAD solution robust to adversarial presentation attacks. 

Competitors can choose whether to participate in the first or second challenge or both. All competitors will participate in the third challenge.

Each participant is invited to submit its algorithm in a Windows or Linux console application. The performance will be evaluated by utilizing a very large data set of “fake” and “live” fingerprint images captured by four devices. The performance rank will be compiled and published in this site.

The goal of the competition is to compare different methodologies for software-based fingerprint liveness detection with a common experimental protocol and data set. The ambition of the competition is to become the reference event for academic and industrial research. The competition is not defined as an official system for quality certification of the proposed solutions, but may impact state-of-art in this crucial field, with reference to the general problem of security in biometric systems.

 

---

IMPORTANT DATES
Deadlines:

• Registration Opening: April 1, 2024.
• Registration Closing:  September 30, 2024.
• Algorithms Submission: December 20, 2024.

Competitors must select state-of-the-art datasets for model training and declare the data used, as no training data will be provided.

 

LivDet 2023 results were presented during the 2023 International Joint Conference on Biometrics (IJCB 2023).

Preprint on: https://arxiv.org/abs/2309.15578

---

^{(1) The “Liveness Detection in Action” name is inspired by “Anti-spoofing in Action: Joint Operation with a Verification System, I. Chingovska, A. Anjos and S. Marcel,  2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops, Portland, OR, 2013, pp. 98-104. doi: 10.1109/CVPRW.2013.22″}

^{(2) Ghiani, L., Marcialis, G.L., Roli, F., Tuveri, P.: User-specific effects in fingerprint presentation attacks detection: insights for future research. In: 2016 ICB, Halmstad, pp. 1–6 (2016). doi: 10.1109/ICB.2016.7550081}

^{(3) Micheletto, M., Marcialis, G. L., Orrù, G., & Roli, F. (2021), Fingerprint recognition with embedded presentation attacks detection: are we ready?, IEEE Transactions on Information Forensics and Security, 16, 5338-5351, doi: 10.1109/TIFS.2021.3121201.}

IEEE Copyright Notice: 
This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author’s copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.