Luisa Verdoliva

Luisa Verdoliva is Full Professor at University Federico II of Naples, Italy, where she leads the Multimedia Forensics Lab. Her scientific interests are in the field of image and video processing, with main contributions in the area of multimedia forensics. She is the PI for University Federico II of Naples in the DISCOVER (a Data-driven Integrated Approach for Semantic Inconsistencies Verification) project funded by DARPA.

Keynote abstract - Can Forensic Detectors Keep Up with the Rise of Generative AI?

With the rapid progress of recent years, techniques that generate and manipulate multimedia content now reach a very advanced level of realism. Powered by large language models, text-driven synthesis tools allow the user to modify or create from scratch images and videos by means of simple text instructions. Such rapid technological developments on the generation side require that detectors also evolve continuously and effectively to keep ensuring high reliability. This talk analyzes the trustworthiness of current deepfake detectors and their limitations in real-world scenarios, finally presenting some possible strategies to overcome them.