Chameleon: On the Scene Diversity and Domain Variety of AI-Generated Videos Detection
This work addresses the problem of limited forensic tools for AI-generated video detection, which is crucial for combating disinformation, though it is incremental as it focuses on dataset construction rather than a new detection method.
The paper tackled the lack of diverse and realistic datasets for detecting AI-generated videos by constructing Chameleon, a dataset that includes multiple generation tools, real video sources, and complex elements like scene switches and dynamic perspective changes, expanding detection beyond faces to human actions and environments.
Artificial intelligence generated content (AIGC), known as DeepFakes, has emerged as a growing concern because it is being utilized as a tool for spreading disinformation. While much research exists on identifying AI-generated text and images, research on detecting AI-generated videos is limited. Existing datasets for AI-generated videos detection exhibit limitations in terms of diversity, complexity, and realism. To address these issues, this paper focuses on AI-generated videos detection and constructs a diverse dataset named Chameleon. We generate videos through multiple generation tools and various real video sources. At the same time, we preserve the videos' real-world complexity, including scene switches and dynamic perspective changes, and expand beyond face-centered detection to include human actions and environment generation. Our work bridges the gap between AI-generated dataset construction and real-world forensic needs, offering a valuable benchmark to counteract the evolving threats of AI-generated content.