Abstract
The paper presents the computer system for detecting deep fake images in video films. The system is based on applyingcontinuous wavelet transformation combined with the ensemble of classifiers composed of a few convolutional neural networks of diversified architecture. Three different forms of forged images taken from the Face-Forensics++ database are considered in numerical experiments. The results of experiments on the application of the proposed system have shown good performance in comparison to other actual approaches to this problem.
Publisher
National Institute of Telecommunications
Subject
Electrical and Electronic Engineering,Computer Networks and Communications
Reference23 articles.
1. A. Rossler et al., "FaceForensics++: Learning to Detect Manipulated Facial Images", in: Proceedings of the IEEE International Conference on Computer Vision (ICCV), Seoul, South Korea, 2019.
2. L. Jaing, R. Li, W. Wu, C. Qian, and C.C. Loy, "Deeperforensics-1.0: a Large-scale Data Set for Real-world Face Forgery Detection", 2020.
3. P. Yu, Z. Xia, J. Fei, and Y. Lu, "A Survey on Deepfake Video Detection", IET-Biometrics, vol. 10, no. 6, pp. 607-624, 2021.
4. D. Cozzolino, G. Poggi, and L. Verdoliva, "Extracting Camera-based Finger Prints for Video Forensics", in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, Long Beach, USA, 2019.
5. H.H. Nguyen, J. Yamagishi, and I. Echizen, "Capsule-forensics: Using Capsule Networks to Detect Forged Images and Videos", in: 2019 IEEE International Conference on Acoustics, Speech and Signal Processing, Brighton, UK, pp. 2307-2311, 2019.