摘要
数字媒体的迅速普及和易于操纵，需要强大的伪造检测技术来保持多媒体的可信度。这篇评论文章全面概述了过去十年伪造检测技术的进步，重点介绍了传统方法、基于机器学习的方法和基于深度学习的方法。传统技术涉及水印、签名和统计属性分析，而基于机器学习的方法则采用监督学习进行自动伪造分类。基于深度学习的方法利用卷积神经网络（CNN）从原始像素数据中学习分层特征，在检测高级操纵方面表现出色。尽管取得了这些进步，但挑战依然存在，包括标记数据的可用性有限、对抗性攻击、跨不同伪造技术的泛化以及实时检测。应对这些挑战对于提高数字媒体的可信度和维护数字景观的完整性至关重要。这篇评论文章旨在全面了解多媒体伪造检测的现状，并启发未来的研究方向以应对剩余的挑战。
关键词： 伪造检测；多媒体；机器学习；深度学习；数字媒体
Abstract
The rapid proliferation of digital media and ease of manipulation necessitate robust forgery detection techniques to maintain multimedia trustworthiness. This review paper offers a comprehensive overview of the advancements in forgery detection techniques over the past decade, focusing on traditional, machine learning-based, and deep learning-based approaches. Traditional techniques involve watermarking, signatures, and statistical property analysis, while machine learning-based methods employ supervised learning for automatic forgery classification. Deep learning-based methods utilize convolutional neural networks (CNNs) to learn hierarchical features from raw pixel data, demonstrating exceptional performance in detecting advanced manipulations. Despite these advancements, challenges persist, including limited availability of labeled data, adversarial attacks, generalization across different forgery techniques, and real-time detection. Addressing these challenges is crucial for enhancing the trustworthiness of digital media and preserving the integrity of the digital landscape. This review paper aims to provide a thorough understanding of the current state of multiple-media forgery detection and inspire future research directions to tackle remaining challenges.
Key words： Forgery Detection; Multiple-media; Machine learning; Deep learning; Digital Media
参考文献 References
[1] Farid, H. (2009). Digital image forensics. IEEE Signal Processing Magazine, 26(2), 26-37.
[2] Afchar, D., Nozick, V., Yamagishi, J., & Echizen, I. (2018). Mesonet: a compact facial video forgery detection network. IEEE Transactions on Information Forensics and Security, 13(5), 1211-1226.
[3] Bayram, S., Sencar, H. T., & Memon, N. (2010). An efficient and robust method for detecting copy-move forgery. In 2010 IEEE International Conference on Acoustics, Speech and Signal Processing, 1053-1056. IEEE.
[4] Zhang, J., Han, B., Lin, Z., & Yang, M. (2018). Image manipulation detection using semi-supervised learning. In Proceedings of the European Conference on Computer Vision (ECCV), 802-816.
[5] Rossler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., & Nießner, M. (2019). Faceforensics++: Learning to detect manipulated facial images. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 1-11.
[6] LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.
[7] Li, Y., Chang, M. C., & Lyu, S. (2020). In ictu oculi: Exposing AI-generated fake face videos by detecting eye blinking. IEEE Transactions on Information Forensics and Security, 15, 2316-2325.
[8] Nguyen, H. H., Yamagishi, J., & Echizen, I. (2019). Capsule-forensics: Using capsule networks to detect forged images and videos. In 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2307-2311. IEEE.
[9] Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014). Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572.
[10] Agarwal, S., Farid, H., Gu, Y., & He, M. (2021). Protecting world leaders against deep fakes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2052-2061.
[11] Zhou, P., Han, X., Morariu, V. I., & Davis, L. S. (2021). Learning rich features for image manipulation detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1054-1063.