Vai al contenuto

Biometrics

PRA Lab’s Biometric Unit was founded by Prof. Gian Luca Marcialis, who is currently head of the research group, since 2001. The PRA Lab’s Biometric Unit has gained a significant experience in fingerprints and faces recognition systems, detection of presentation attacks, deepfakes, and anomalous events in crowds.

RESEARCH TOPIC

Personal authentication

Personal recognition through physiological and behavioral characteristics.

RESEARCH TOPIC

Presentation attacks

Protect biometric systems from attacks targeting the sensor.

RESEARCH TOPIC

DeepFake detection

Detection of facial identities manipulations in video and images.

RESEARCH TOPIC

Crowd Analysis

Anomaly detection in crowd analysis refers to the ability to detect events that deviate from normality.

RESEARCH TOPIC

Multimodal biometrics

Combining information coming from different biometric traits and soft-biometrics.

RESEARCH TOPIC

Behavioral biometrics

Analysis of behavior and biomedical signals in the biometric domain.

People involved

Fabio
Roli

Full Professor

Gian Luca
Marcialis

Associate Professor

Luca
Didaci

Associate Professor

Giulia
Orrù

RTDa

Marco
Micheletto

RTDa

Roberto
Casula

PostDoc

Sara
Concas

PhD student

Simone Maurizio
La Cava

PhD student

Gianpaolo
Perelli

PhD student

Active projects

BullyBuster – A framework for bullying and cyberbullying action detection by computer vision and artificial intelligence methods and algorithms

Supported by the Italian Ministry of Education, University and Research (MIUR) within the PRIN2017, the project has been included in the Global Top 100 list of AI projects addressing the 17 UNS-DGs (United Nations Strategic Development Goals) by the International Research Center for Artificial Intelligence under the auspices of UNESCO.

Fingerprint Liveness Detection Competition

In order to cover the lack of common data and protocols to assess the performance of fingerprint presentation attack detection (PAD) systems, the Biometric Unit organized the International Fingerprint Liveness Detection Competition (LivDet) since from 2009. The LivDet competitions aim to assess the performance of fingerprint PAD algorithms using a common experimental protocol and data sets. The competition is open to academic and industrial institutions which have developed fingerprint PAD algorithms.