Keynote speakers from: 2025 | 2024 | 2023 | 2022 | 2021 | 2020 | 2019 | 2018 | 2017 | 2016 | 2015 | 2014 | 2013 | 2012 | 2011

Marco Porta
Professor, University of Pavia, Italy

Keynote on: Intelligent Interaction at a Glance: The Future of Gaze-Aware Systems
View Abstract

Biography: Marco Porta is a full professor at the University of Pavia (Pavia, Italy). He has a Master's degree in Electronic Engineering from the Polytechnic of Milan (Italy) and a Ph.D. in Electronic and Computer Engineering from the University of Pavia. He is a member of the Computer Vision & Multimedia Lab research group of the Department of Electrical, Computer and Biomedical Engineering. His main research interests include eye tracking, vision-based perceptive interfaces, visual languages and communication, e-learning, and human-computer interaction in general, with an increasing focus on the use of artificial intelligence and machine perception for human-centered system design. He is also president of the teaching council of the interdepartmental programs CIM (Communication, Innovation, Multimedia, bachelor) and CoD (Digital Communication, master) of the University of Pavia, and (since January 2026) chair of the IEEE Industrial Electronics Society's Technical Committee on Factory Automation.

Houcemeddine Hermassi
Associate Professor, Head of Computer Engineering Department, National School of Engineers of Carthage ENICarthage, University of Carthage

Keynote on: STEG-Guard: A Self-Supervised Multi-Domain Graph Neural Network Framework for Zero-Day Malware Detection in Steganographic Images
View Abstract

Biography: Houcemeddine Hermassi is an Associate Professor and Head of the Computer Engineering Department at the National Engineering School of Carthage (ENICarthage), University of Carthage, Tunisia. He is also a senior researcher at the RISC Laboratory of the National School of Engineers of Tunis (ENIT-University Elmanar). He holds a PhD in Telecommunications from ENIT, Tunisia, with a dissertation focused on cryptology of multimedia content based on chaos theory and DNA computing. He also holds a Master of Science in Communication Systems from ENIT-University Elmanar and a National Diploma of Engineer in Communications and Networks. His research interests include data encryption, steganography, digital watermarking, intrusion detection using artificial intelligence, and AI-based cryptology. He has contributed to several national and international research projects and is the author and co-author of more than 30 scientific publications in high-impact peer-reviewed journals and international conference proceedings. His work lies at the intersection of cybersecurity, artificial intelligence, and multimedia information security, with a focus on developing advanced techniques for secure communication and intelligent threat detection systems.

Intelligent Interaction at a Glance: The Future of Gaze-Aware Systems
Marco Porta

Abstract: As computing systems evolve toward more natural and adaptive interaction, gaze offers a powerful channel for both implicit understanding and explicit control. The presentation explores how modern eye tracking, combined with machine learning and contextual intelligence, enables systems that infer user attention and respond to gaze as an explicit input modality. From hands-free interaction to adaptive visual systems, the talk examines applications, design challenges, and implications of gaze-aware technology for the future of human-computer interaction.

STEG-Guard: A Self-Supervised Multi-Domain Graph Neural Network Framework for Zero-Day Malware Detection in Steganographic Images
Houcemeddine Hermassi

Abstract: Malware concealed within images via steganography poses a significant challenge for traditional intrusion detection systems, as it evades standard signature- and behavior-based detection methods. This paper introduces STEG-Guard, a novel AI-driven framework that combines multi-domain feature extraction, graph neural networks (GNNs), and self-supervised learning to detect malware hidden in images. Each image is represented as a heterogeneous graph integrating spatial, frequency, and noise-residual information, capturing subtle perturbations introduced by steganographic embedding. A hybrid CNN–GNN encoder fuses global and local representations, enabling robust detection without reliance on labeled datasets. Experimental evaluation on benchmark steganography datasets augmented with real and synthetic malware payloads demonstrates that STEG-Guard achieves high accuracy and strong zero-day detection capabilities, outperforming state-of-the-art CNN and statistical methods. The framework also provides a foundation for integrating image-based steganalysis into network intrusion detection systems, offering a scalable and generalizable approach to combating hidden malware threats.