The Workshop MultiMedia FORensics in the WILD (MMForWILD) 2024 will be held on December 1st 2024, from 9:30 to 14:00 (Pacific Time -PDT).
Each technical presentation will run for 15 minutes, followed by a 5-minute Q&A session. It will be conducted in a hybrid format, allowing participants to join either in person or online. The online sessions will be accessible via the following MS Teams link: MS Teams Link
The schedule of the workshop will be as follows:
9:00 – 9:15 - Opening & Welcome by Chair
9:15 – 10:15
Invited Speaker: Luca Guarnera
Title: Deepfake Technology: Insights into Creation and Detection
Abstract: A comprehensive overview of the deepfake phenomenon, examining its rapid development and the dual challenges it presents in both creation and detection. This introduction explores the technologies enabling deepfake generation, discussing applications and potential risks across various domains. Current detection techniques are reviewed, highlighting ongoing research and advancements aimed at identifying and mitigating deepfake content. By analyzing both the creative and defensive aspects of deepfakes, this overview offers insights into the evolving landscape of digital media manipulation.
Technical Session: 10:15 - 13:00
- 10:15 - 10:35 Supriyo Sadhya and Xiaojun Qi - Detection of Deepfakes Using Wavelet-Based Convolutional Neural Network (online)
- 10:35 - 10:55 Stefano Borzì, Lorenzo Mongelli, Filippo Stanco, Sebastiano Battiato and Dario Allegra - Breaking the Silence: Detecting AI-Converted Voices in the Quietest Moments
- Break: 11:00 - 11:30
- 11:30 - 11:50 Andreas Vorderleitner, Jutta Hämmerle-Uhl and Andreas Uhl - Finger and Hand Vein Spoof GANs: Assessment of Extended Presentation Attack Instrument Sample Sets (online)
- 11:50 - 12:20 Luca Guarnera, Francesco Guarnera, Alessandro Ortis, Sebastiano Battiato and Giovanni Puglisi - Evasion Attack on Deepfake Detection via DCT Trace Manipulation
- 12:20 - 12:40 Mehdi Atamna, Iuliia Tkachenko and Serge Miguet - WaveConViT: Wavelet-Based Convolutional Vision Transformer for Cross-Manipulation Deepfake Video Detection
- 12:40 - 13:00 Hessen Bougueffa Eutamene, Wassim Hamidouche, Mamadou Keita, Abdelmalik Taleb-Ahmed and Abdenour Hadid -Synthetic Image Detection using Mixture of Knowledge Distillation from Vision-Language Models
- 13:00 - Closing Remarks by Chair