CoChat

Deepfake Video Detection via Predictive Representation Learning

Open access
Ge, Shiming, Lin, Fanzhao, Li, Chenyu, Zhang, Daichi, Wang, Weiping, Zeng, Dan · 2022 · ACM Transactions on Multimedia Computing, Communications, and Applications
Cited by 64

The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.

Abstract

Increasingly advanced deepfake approaches have made the detection of deepfake videos very challenging. We observe that the general deepfake videos often exhibit appearance-level temporal inconsistencies in some facial components between frames, resulting in discriminative spatiotemporal latent patterns among semantic-level feature maps. Inspired by this finding, we propose a predictive representative learning approach termed Latent Pattern Sensing to capture these semantic change characteristics for deepfake video detection. The approach cascades a Convolution Neural Network-based encoder, a ConvGRU-based aggregator, and a single-layer binary classifier. The encoder and aggregator are pretrained in a self-supervised manner to form the representative spatiotemporal context features. Then, the classifier is trained to classify the context features, distinguishing fake videos from real ones. Finally, we propose a selective self-distillation fine-tuning method to further improve the robustness and performance of the detector. In this manner, the extracted features can simultaneously describe the latent patterns of videos across frames spatially and temporally in a unified way, leading to an effective and robust deepfake video detector. Extensive experiments and comprehensive analysis prove the effectiveness of our approach, e.g., achieving a very highest Area Under Curve (AUC) score of 99.94% on FaceForensics++ benchmark and surpassing 12 states of the art at least 7.90%@AUC and 8.69%@AUC on challenging DFDC and Celeb-DF(v2) benchmarks, respectively.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1145/3536426

Cites

26 of 26 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 34

  1. 10.1109/CVPRW50498.2020.00338 · DOI · Paper page
  2. 10.1109/CVPR.2016.572 · DOI · Paper page
  3. Nicolas Ballas, Li Yao, Chris Pal, and Aaron Courville. 2016. Delving deeper into convolutional networks for learning video representations. In Proceedings of the International Conference on Learning Representations (ICLR’16).
  4. 10.1109/TMM.2020.2980194 · DOI
  5. 10.5555/184656.180369 · doi:10.5555/184656.180369
  6. 10.1109/CVPR.2017.502 · DOI · Paper page
  7. 10.1145/2070781.2024164 · DOI · Paper page
  8. Emily Denton and Vighnesh Birodkar. 2017. Unsupervised learning of disentangled representations from video. In Proceedings of the International Conference on Neural Information Processing Systems (NeurIPS’17). 4417–4426.
  9. Jean-Yves Franceschi, Edouard Delasalles, Mickaël Chen, Sylvain Lamprier, and Patrick Gallinari. 2020. Stochastic latent residual video prediction. In Proceedings of the International Conference on Machine Learning (ICML’20). 3233–3246.
  10. 10.1109/TIP.2018.2883743 · DOI · Paper page
  11. Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. In Proceedings of the International Conference on Neural Information Processing Systems (NeurIPS’14). 2672–2680.
  12. 10.1109/AVSS.2018.8639163 · DOI · Paper page
  13. 10.1109/ICCVW.2019.00186 · DOI · Paper page
  14. 10.1007/978-3-030-58580-8_19 · DOI · Paper page
  15. Javier Hernandez-Ortega, Rubén Tolosana, Julian Fiérrez, and Aythami Morales. 2021. Deepfakeson-phys: Deepfakes detection based on heart rate estimation. In Proceedings of the AAAI Conference on Artificial Intelligence Workshop on Artificial Intelligence Safety (AAAIW’21). 1–8.
  16. Geoffrey Hinton, Oriol Vinyals, and Jeffrey Dean. 2015. Distilling the knowledge in a neural network. In Proceedings of the International Conference on Neural Information Processing Systems (NeurIPS’15) Workshop. 1–9.
  17. Sohail Ahmed Khan, Alessandro Artusi, and Hang Dai. 2021. Adversarially robust deepfake media detection using fused convolutional neural network predictions. arXiv:2102.05950. Retrieved from https://arxiv.org/abs/2102.05950. · arXiv
  18. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In Proceedings of the International Conference on Learning Representations (ICLR’15).
  19. Yuezun Li, Xin Yang, Pu Sun, Honggang Qi, and Siwei Lyu. 2019. Celeb-DF (v2): A new dataset for deepfake forensics. arXiv:1909.12962. Retrieved from https://arxiv.org/abs/1909.12962. · arXiv
  20. Ruofan Liang, Tianlin Li, Longfei Li, Jing Wang, and Quanshi Zhang. 2020. Knowledge consistency between neural networks and beyond. In Proceedings of the International Conference on Learning Representations (ICLR’20).
  21. 10.1145/3524497 · DOI · Paper page
  22. 10.5244/C.29.41 · DOI · Paper page
  23. Ekraam Sabir, Jiaxin Cheng, Ayush Jaiswal, Wael AbdAlmageed, Iacopo Masi, and Prem Natarajan. 2019. Recurrent convolutional strategies for face manipulation detection in videos. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW’19). 80–87.
  24. 10.1109/CVPR.2015.7298682 · DOI · Paper page
  25. Chen Sun, Fabien Baradel, Kevin Murphy, and Cordelia Schmid. 2019. Learning video representations using contrastive bidirectional transformer. arXiv:1906.05743. Retrieved from https://arxiv.org/abs/1906.05743. · arXiv
  26. 10.1109/CVPR46437.2021.00361 · DOI · Paper page
  27. 10.1145/3306346.3323035 · DOI · Paper page
  28. 10.1145/2816795.2818056 · DOI · Paper page
  29. 10.1145/2929464.2929475 · DOI · Paper page
  30. Laurens Van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. J. Mach. Learn. Res. 9, 86 (2008), 2579–2605.
  31. 10.1145/3474085.3478324 · DOI · Paper page
  32. 10.1109/FG47880.2020.00089 · DOI · Paper page
  33. Natalie Wolchover and Lucy Reading. 2017. New theory cracks open the black box of deep learning. Quanta Mag. 3 (2017).
  34. 10.1109/WACVW50321.2020.9096939 · DOI · Paper page

Cited by

See all 64

Metadata

Authors
Ge, S., Lin, F., Li, C., Zhang, D., Wang, W., & Zeng, D.
Venue
ACM Transactions on Multimedia Computing, Communications, and Applications
Published
2022-06-30
Volume
18
Issue
2s
Pages
1-21
ISSN
  • 1551-6857 (ISSN-L)
  • 1551-6865
OpenAlex
W4280579728
Cited by
64
References
26
FWCI
4.21
Type
journal article
Language
en
Open access
hybrid
Funders
  • National Key Research and Development Plan (2020AAA0140001)
  • Beijing Natural Science Foundation (19L2040)
  • National Natural Science Foundation of China (61772513)
  • Youth Innovation Promotion Association, Chinese Academy of Sciences

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.