Multimedia Signal Processing and Understanding Lab

The Multimedia Signal Processing and Understanding Lab (mmlab) deals with relevant aspects of multimedia data processing, focusing on both theoretical and application-driven research issues

Since media can be considered as the technological extensions of human capabilities, particular attention is paid to smart multimedia data management, analysis, transmission, authentication and protection, and to those applications where ambient intelligence can provide advanced services beyond human limitations

What's going on

Giulia Boato and Francesco De Natale talking on multimedia credibility assessment, fake images and videos, and forensics technologies:

https://www.ictsecuritymagazine.com/articoli/credi-a-cio-che-vedi-il-sempre-piu-sottile-confine-fra-reale-e-virtuale/

Source: StyleGAN

Ongoing online study

Recent AI technologies can create from scratch stunningly realistic images of faces that do not belong to any real human. To which extent are we still able to spot them at a glance?

We have developed an online perceptual study with the goal of assessing people ability to distinguish between real and synthetic face images when confronted with cutting-edge technologies.

All individuals who are at least 18 years old are welcome to participate.

This study is conducted by the MMLab (University of Trento - Italy) in collaboration with the GoffauxLab (Catholic University of Louvain - Belgium) and the Security & Privacy Lab (University of Innsbruck - Austria).