Identifying Rhythmic Patterns for Face Forgery Detection and Categorization

TL;DR

This paper introduces a novel framework leveraging rhythmic PPG signal patterns to detect and categorize face forgeries, demonstrating superior performance through extensive experiments.

Contribution

The paper presents a new rhythmic pattern-based approach for face forgery detection and categorization, utilizing specialized neural networks and blending techniques.

Findings

Effective detection and categorization of face forgeries.

Superior performance over existing methods.

Robustness across different forgery sources.

Abstract

With the emergence of GAN, face forgery technologies have been heavily abused. Achieving accurate face forgery detection is imminent. Inspired by remote photoplethysmography (rPPG) that PPG signal corresponds to the periodic change of skin color caused by heartbeat in face videos, we observe that despite the inevitable loss of PPG signal during the forgery process, there is still a mixture of PPG signals in the forgery video with a unique rhythmic pattern depending on its generation method. Motivated by this key observation, we propose a framework for face forgery detection and categorization consisting of: 1) a Spatial-Temporal Filtering Network (STFNet) for PPG signals filtering, and 2) a Spatial-Temporal Interaction Network (STINet) for constraint and interaction of PPG signals. Moreover, with insight into the generation of forgery methods, we further propose intra-source and inter-source blending to boost the performance of the framework. Overall, extensive experiments have proved the superiority of our method.