Multimedia Forensics

The course aims to provide advanced knowledge in the field of Multimedia Forensics, with a particular focus on the analysis, recognition, and detection of digital manipulations (images, video, audio, and multimodal content), as well as on the regulatory and ethical aspects related to deepfakes and anti-forensic techniques.

Knowledge and understanding:
Students will acquire a solid understanding of the fundamental principles of Multimedia Forensics, of the main techniques for the acquisition and manipulation of multimedia content, as well as of the methodologies for the detection and attribution of alterations. Key concepts such as digital forgery, deepfakes, machine learning and deep learning for detection, multimodal approaches, and the challenges posed by adversarial machine learning will be explored in depth.

Applying knowledge and understanding:
Through laboratory sessions and case studies, students will be able to apply the acquired methodologies to identify and analyze digital manipulations in images, video, and audio.

Making judgements:
By being exposed to a variety of forgery and detection scenarios, students will develop the ability to critically assess the reliability of the obtained results, recognize the limitations of current techniques, and propose innovative solutions. They will also be able to reflect on the ethical, legal, and social implications of deepfakes and anti-forensic techniques.

Communication skills:
Students will gain communication skills and mastery of the technical-scientific language in the field of Multimedia Forensics, with particular attention to technical, legal, and investigative terminology.

Learning skills:
The course will provide students with the theoretical and practical tools to independently keep up with the rapid developments of the field, fostering a critical approach and a problem-solving oriented working methodology. Thematic seminars and advanced laboratories will further stimulate the ability to learn new methods and adapt to emerging scenarios in digital and media forensics.

• Module 1: Fundamentals of Digital & Image Forensics
• Module 2: Digital Forgery, JPEG Forensics, and JPEG AI
• Module 3: Deepfakes, Introduction to Generative Models, and Synthetic Data Creation
• Module 4: Deepfake Detection and Attribution in Images and Videos
• Module 5: Impostor Bias and Deepfake Regulation
• Module 6: Audio Deepfakes: Creation and Detection Techniques
• Module 7: Multimodal Deepfake Detection
• Module 8: Adversarial Machine Learning and Emerging Challenges