Roberto Giordano
• Alessio Pittiglio
sommario
Deepfake technology, which uses artificial intelligence to generate fake videos,
presents significant ethical and social challenges. These manipulated videos
can be used to spread disinformation, undermining public trust in the media.
This project aims to explore deepfake detection methods by analyzing vari-
ous approaches. We will start with well-established CNN-based baselines (e.g.,
Xception [1]) and may also investigate more recent solutions, applying them
to publicly available datasets. Based on this analysis, we will develop a pro-
totype system capable of detecting whether a video is real or not. The goal
is to contribute to addressing the ethical challenges in AI, such as promoting
transparency and preventing the malicious use of emerging technologies.
prodotti