TWYN: Trust What You learN, Workshop on Trustworthiness in Computer Vision at ECCV 2024
On September 30, 2024, the first Trust What You learN (TWYN) workshop on trustworthiness in computer vision took place at the MiCo congress center in Milan, in conjunction with ECCV 2024, with the support of the ELSA project.
The workshop brought together researchers, practitioners and industry experts at the intersection of Trustworthy AI and DeepFake Analysis, around two tracks: From Learning to Unlearning: the Role of Privacy in Computer Vision, on differential privacy, privacy attacks, machine unlearning and privacy-preserving learning, and DeepFake Analysis and Detection, on methods to distinguish real from generated or altered visual content. Accepted papers were presented in a poster session.
Keynote talks
| 09:05 | Can Forensic Detectors Keep Up with the Rise of Generative AI?Luisa Verdoliva, University of Naples Federico II |
| 09:30 | Deep Fakes: quo vadis?Cristian Canton Ferrer, Meta AI |
| 10:00 | Trustworthy End-to-End Document UnderstandingDimosthenis Karatzas, Universitat Autònoma de Barcelona |
| 10:30 | Coffee break and poster session |
| 11:00 | Memorization in Machine LearningReza Shokri, National University of Singapore |
| 11:30 | Can a Neural Network Understand What’s Real?Stefano Soatto, University of California, Los Angeles |
| 12:00 | Schizophrenic Struggles in AI: Balancing Learning and Unlearning in Fake GenerationRita Cucchiara, University of Modena and Reggio Emilia |
Organizers
Marco Cotogni, Luigi Sabetta and Jacopo Bonato (Leonardo SpA), Sara Sarto, Samuele Poppi and Lorenzo Baraldi (University of Modena and Reggio Emilia).
An archived copy of the workshop website is available.