International Summer School on Biometrics, Behavior and Vision

BSS

Apply now

Syllabus

Course Programme

A 40-hour programme across five days. Topics and sessions are tailored to each edition; the final daily programme is awaiting confirmation.

Apply now
Download schedule

Summer school focus

"AI Agents for Biometrics, Behavior & Vision"

Over the last decade the advent of Deep Learning and, more recently, Agentic AI, strongly impacted both research and application of automatic biometric data analysis and visual recognition. However, other issues and concerns, not directly related to the computational performance, arise including the fairness, reliability and trustworthiness of current biometric and vision systems.

Questions that will be answered:

This edition of the school will address these issues by trying to answer some compelling questions, such as:

  • The impact of AI and advanced learning techniques in Biometrics.

  • What can we learn from human perception?

  • How to better deploy current large foundation models?

  • How human interactions can facilitate continual learning and adaptation?

  • What is the potential impact of biometrics in forensic investigation and crime prevention?

The school lectures will cover all main aspects of advanced biometric technologies, including a view of several modalities and integration strategies:

  • Gait and posture.

  • Fingerprints;

  • Human face;

  • Iris and periocular;

  • Soft biometrics;

  • Speech.

The lectures will also cover the most advanced issues in the application of Deep Learning to biometric technologies, including:

  • AI Foundation Models;

  • Multimedia Generative Models;

  • Biometric challenges and competitions;

  • Collection and management of large-scale datasets;

  • Detection and resolution of morphing attacks;

  • Ethical, legal and social aspects;

  • Evaluation and assessment strategies;

  • Forensic applications;

  • Security and data protection.

Some lectures will be devoted to multidisciplinary subjects, allowing participants to better understand basic and side issues in the development of biometric systems, including:

  • Commonalities of deep CNNs and neural architectures;

  • Explainable AI;

  • Federated learning;

  • Machine learning (deep and shallow);

  • Neural systems for perception;

  • Structure and functionality of biological neural architectures.

Academic Blueprint

Five-Day Programme

Topics change with each edition. Download the draft programme below.

The classes will be held during the week in two sessions (morning and afternoon) with a coffee break in the middle of each session. The course timings will be arranged to allow some leisure and free time for the students to take advantage of the beautiful environment.

A special session will be held to allow the students to present their recent research outcomes related to biometrics.

All participating Phd students and postdocs are warmly invited to discuss a focused research project with one of the lecturers to be jointly developed.

Flipbook Traditional Download

Learning Mechanics

School format & activities

Beyond traditional lecturing, the school features structural avenues for participant integration.

60 PhD candidates participate in peer review tasks, deliver lightning talks on thesis hypotheses, and present original research on poster boards evaluated by the steering committee.

Download schedule

Admissions are highly selective

Join hand-picked researchers and practitioners in Alghero for five intensive days.

Apply now