Data-Driven Deepfake Forensics Model Based on Large-Scale Frequency and Noise Features

With the rapid development of deep learning and communication technology, the application of streaming media services and social software have gone deep into life. However, in the face of many uncertain factors in data dissemination, protecting privacy and security is particularly important. In order to solve the abovementioned problems, this study proposes a deep face forgery forensics method with frequency domain and noise features. In this method, discrete cosine transform is proposed to perceive the forgery trace features of different frequency bands in the frequency domain. At the same time, the spatial rich model is used for guidance to enhance the traces of forged noise. Then, large-scale network and single center loss function are introduced to improve the forensics ability of the model. Experimental results on several databases such as faceforensics++, celeb DF, and DFDC show that this method can effectively improve the accuracy of forensics.

Authors

• Guipeng Lan Tianjin Unversity
• Shuai Xiao Tianjin Univesity
• Jiabao Wen Tianjin University
• Desheng Chen
• Yong Zhu

Keywords

• Feature extraction
• Forgery
• Training
• Frequency-domain analysis
• Convolutional neural networks
• Face recognition
• Forensics
• Noise measurement
• Streaming media
• Privacy
• Security
• Discrete cosine transforms
• Noise Characteristics
• Forensic Models
• Loss Function
• Deep Learning
• Frequency Band
• Single Center
• Convolutional Layers
• Frequency Domain
• News Media
• Human Eye
• Generative Adversarial Networks
• Single Function
• Domain Features
• Frequency Noise
• Single Training
• Metric Learning
• Technological Revolution
• Discrete Cosine Transform
• Frequency Domain Features
• Training Loss Function
• Frequency Domain Information
• Backbone Network
• Dual Model
• Convolutional Neural Network
• Real Faces
• Spatial Features
• Central Point
• Image Block
• Convolutional Network
• Dual Mode