AI-Driven Multimodal Biometrics for Enhanced Personal Authentication

Biometric technologies have rapidly transitioned from laboratory prototypes to mission-critical infrastructure that enables seamless authentication, identity verification, and access control across consumer, enterprise, and public sector applications. Modern systems increasingly rely on multimodal fusion, combining fingerprints, face, iris, voice, gait, vein, and behavioral signals to boost accuracy, reduce failure rates, and improve resilience to environmental and demographic variability. In parallel, soft biometrics (e.g., age, gender, emotion) are often used to enrich recognition pipelines and retrieval, but they also raise important questions around fairness, privacy, and social impact.

As these systems scale, they face new threats: presentation attacks/spoofing, impersonation and deepfakes, adversarial examples, model extraction, and template inversion. Building truly trustworthy biometrics therefore requires advances in liveness detection, robust and explainable AI, privacy-preserving learning (e.g., federated and encrypted computation), and rigorous evaluation protocols. Equally crucial are hardware–software co-design (from edge sensors and TinyML to secure enclaves), dataset curation with strong governance, and standardized benchmarks that reflect real-world distribution shifts and open-set conditions.

This Special Issue invites contributions that push the state of the art while foregrounding responsible deployment. We particularly seek work that connects algorithmic innovation to system-level performance, reproducibility, and legal/ethical compliance (e.g., transparency, accountability, data protection). We welcome original research, surveys and tutorials, datasets/benchmarks, and case studies documenting deployments “in the wild.” By convening researchers across machine learning, security, HCI, law, and ethics, this Special Issue aims to chart a principled path for multimodal biometrics that is accurate, robust to adversaries, privacy-aware, fair across populations, and compliant with evolving regulatory landscapes.

Topics of interest include (but are not limited to) the following:

• Fingerprint, face, iris, periocular, voice/speaker, gait, vein, keystroke, and other behavioral biometrics;
• Soft biometrics (age, gender, emotion, attributes) and their ethical/privacy implications;
• Verification/identification at scale, open-set recognition, watchlist, and de-duplication;
• Presentation attack detection (PAD), spoofing and deepfake detection, liveness cues, and sensor fusion;
• Adversarial attacks and defenses for biometric pipelines, certified robustness, adversarial training, and red-teaming;
• Template security: cancelable biometrics, biometric cryptosystems, differential privacy, and secure enclaves/TEE;
• Privacy-preserving and distributed learning: federated, homomorphic encryption, and secure MPC;
• Domain shift, cross-sensor/cross-demographic generalization, and calibration and uncertainty estimation;
• Hardware–software co-design: edge sensors, on-device/TinyML inference, and energy/latency optimization;
• Data quality assessment, dataset governance, protocols, benchmarks, synthetic data, and generative models;
• Explainability, auditability, fairness and bias mitigation, and human-in-the-loop evaluation;
• Standards and regulatory/compliance perspectives (e.g., governance, risk assessment, documentation);
• Surveys, tutorials, datasets, and benchmarks, as well as reproducibility reports and real-world case studies.

Dr. Stefano Marrone
Dr. Roberto Casula
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. AI is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• Multimodal biometrics
• Soft biometrics
• Presentation attack detection
• Liveness detection
• Adversarial machine learning
• Privacy-preserving learning
• Fairness and bias mitigation
• Edge/On-device AI
• Datasets and benchmarks
• Legal and ethical compliance