DFAD 2023: Workshop and Challenge on DeepFake Analysis and Detection at ICCV 2023
On October 2, 2023, the first Workshop and Challenge on DeepFake Analysis and Detection (DFAD) took place in Paris, in conjunction with ICCV 2023. The workshop was organized within the ELSA project, the European Lighthouse on Secure and Safe AI, which builds on the ELLIS network.
DFAD focuses on benchmarks and tools for understanding and detecting fake data, with the goal of protecting against visual disinformation and the misuse of generated images and text. Alongside the paper track, the workshop hosted the ELSA Challenge on DeepFake Detection, run on the ELSA DeepFake dataset collected by the University of Modena and Reggio Emilia and Leonardo SpA. The programme included the presentations of the top challenge entries, five oral papers, a poster session and three invited talks.
Invited talks
| 11:10 | An AI Whodunnit: Following Image Manipulation Clues to their SourceTal Hassner (Meta AI) |
| 11:45 | Sustainable DeepFake Detection, Watermarking, and Personalized DisinformationMario Fritz (CISPA Helmholtz Center for Information Security) |
| 12:20 | Between Illusion and Reality: Guiding AI to Generate Identity-Preserving ImagesNaomi Ken Korem (Lightricks) |
Organizers
Lorenzo Baraldi and Rita Cucchiara (University of Modena and Reggio Emilia), Alessandro Nicolosi (Leonardo SpA), Dmitry Kangin and Plamen Angelov (Lancaster University), Tamar Glaser (Meta AI).
More information is available on the workshop website.