This paper proposes a method that utilizes head pose estimation as a feature to distinguish between real videos and Deepfake videos. The authors analyze the head pose angles of individuals in the videos to detect inconsistencies introduced by Deepfake manipulation. In the experiments, three head pose estimation methods were employed, along with techniques such as KNN and Dynamic Time Warping for experimental validation. The results ultimately demonstrate the effectiveness of this method in Deepfake detection.
Experts Propose a Method to Effectively Identify DeepFakes 'AI Face Swapping'

站长之家
This article is from AIbase Daily
Welcome to the [AI Daily] column! This is your daily guide to exploring the world of artificial intelligence. Every day, we present you with hot topics in the AI field, focusing on developers, helping you understand technical trends, and learning about innovative AI product applications.