The emergence of deep fake technology


EOI: 10.11242/viva-tech.01.05.MCA_07

Download Full Text here



Citation

Prof. Shreya Bhamare, Riya P. Suvarna, Sayali C. Nachare, "The emergence of deep fake technology", VIVA-IJRI Volume 1, Issue 6, Article MCA_07, pp. 1-6, 2023. Published by Computer Engineering Department, VIVA Institute of Technology, Virar, India.

Abstract

Machine learning techniques are increasing the sophistication of the technology. Deep Learning is also known as Deep Structured Learning, which is part of the wider machine learning family. Deep learning architectures such as Deep Neural Networks, Deep Belief Networks, and Convolutional Neural Networks (CNN) have been used for computer vision and language processing. One such deep learning-based application that has emerged recently is "deep fake".Deep fake is a technique or technology where fake images and videos can be created that are difficult for humans to detect. This article looks at how deep fakes are created and what kinds of algorithms are used in them. This will help people learn about the deep fakes that are created daily and a way to tell what is real and what is not.

Keywords

deep fake, deep learning, face swapping, fake detection, Lip-syncing.

References

  1. Li, Xiaoyong, et al. " The paper titled 'The Emergence of Deepfake Technology: Its Potential Impact and Mitigation Strategies' .The paper spans pages 267 to 281.
  2. Zhou, W., Huang, J., & Lin, Z. (2020). Deep fake detection: A survey. arXiv preprint arXiv:2004.11138.
  3. Nguyen, T. H., Nguyen, T. T., Nguyen, D. T., & Vu, D. Q. (2021). Deepfake detection: A review. Journal of Ambient Intelligence and Humanized Computing, 12(4), 3523-3537.
  4. W. Cho, S. Choi, D. K. Park, I. Shin, and J. Choo, “The paper titled 'Image-to-image Translation via Group-wise Deep Whitening-and-Coloring Transformation' The paper spans multiple pp. 10 639–10 647.
  5. T. Karras, S. Laine, and T. Aila, "The paper titled 'A Style-based Generator Architecture for Generative Adversarial Networks' The paper spans across pages 4401 to 4410."
  6. T. Karras, S. Laine, M. Aittala, J. Hellsten, J. Lehtinen, and T. Aila, “Analyzing and improving the image quality of style gun,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 8110–8119.
  7. Rössler, A., Cozzolino, D., Verdoliva, L., Riess C., Thies J., & Nießner, M. (2018). "The paper titled 'FaceForensics: A Large-Scale Video Dataset for Forgery Detection in Human Faces'. The paper presents a comprehensive dataset for detecting forgery in human faces in video format."
  8. C. Vaccari and A. Chadwick, "The paper titled 'Deepfakes and Disinformation: Exploring the Impact of Synthetic Political Video on Deception, Uncertainty, and Trust in News' was published in the academic journal Social Media + Society. The paper examines the effects of synthetic political videos on deception, uncertainty, and trust in news." January-March 2020: 1–13
  9. N. Gardiner, "The Bachelor's thesis titled 'Facial Re-enactment, Speech Synthesis’. The thesis explores the techniques of facial re-enactment and speech synthesis and the emergence of deep fakes."
  10. S. M. Abrar, K. Chowdhury, J. Lubna, "The paper 'Review on Deep Fake: A Looming Technological Threat' was presented at the 11th International Conference on Computing, Communication, and Networking Technologies (ICCCNT) in 2020. The paper provides a comprehensive overview of deep fakes as a potential technological threat."
  11. Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard. Tensorflow: A system for large-sca{le mac}hine learning. In 12th USENIX Symposium o{n Op}erating Systems Design and Implementation ( OSDI 16), pages 265–283, 2016.
  12. Adrian Bulat and Georgios Tzimiropoulos. "The paper 'How Far Are We from Solving the 2D & 3D Face Alignment Problem?' was presented at the IEEE International Conference on Computer Vision in 2017. The paper discusses the progress made towards solving the problem of 2D and 3D face alignment, and presents a dataset of 230,000 3D facial landmarks. The paper spans across pages 1021 to 1030."
  13. Sarwar, S., Islam, S. S., & Akter, F. (2021). Deepfake: An Overview of its Creation, Detection, and Mitigation Techniques. International Journal of Computer Science and Information Security, 19(7), 66-76.
  14. Deepfakes. Deepfakes. https://github.com/deepfakes/faceswap, 2017.
  15. "The paper 'RetinaFace: Single-stage Dense Face Localization in the Wild' by Jiankang Deng, Jia Guo, Yuxiang Zhou, Jinke Yu, Irene Kotsia, and Stefanos Zafeiriou. The paper introduces a single-stage dense face localization technique called RetinaFace, designed for use in unconstrained environments."arXiv preprint arXiv:1905.00641, 2019.
  16. Li, H., et al. (2021). Deepfake Detection using Dynamic Image Frequency and Attention Map Fusion. IEEE Transactions on Multimedia, 23, 2147-2160. doi: 10.1109/TMM.2021.3060112
  17. Nguyen, T. T., et al. (2021). A Robust Watermarking Method for Deepfake Images and Videos. IEEE Transactions on Information Forensics and Security, 16, 3144-3156. doi: 10.1109/TIFS.2021.3077423
  18. Rossler, A., et al. (2019). "The paper titled 'FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces'. The paper describes the creation of a large-scale video dataset that can be used for detecting forgeries in human faces."
  19. Thies, J., et al. (2016). "The paper titled 'Face2Face: Real-time Face Capture and Reenactment of RGB Videos' introduces a real-time system for capturing and reenacting facial expressions in RGB videos 2387-2395. doi: 10.1109/CVPR.2016.262
  20. Zhou, Z., et al. (2020). Detecting Deepfake Audio with an Adaptive Method. IEEE Transactions on Information Forensics and Security, doi: 10.1109/TIFS.2020.2987587