Adversarially Robust Deepfake Video Detection
The DOI matches its Crossref record; 28 of the 29 references it cites were found in a registry or an index.
Abstract
Fake videos have been in circulation on mainstream media since long. However, with increased popularity of online social networks, it is becoming many times easier to spread such videos and achieve virality. Recent advancements in deep learning has further fuelled this menace as the so called deepfake videos are hard to differentiate from the genuine ones. While deepfake video detection techniques attempt to identify the fake videos from real videos, these are now being subjected to adversarial attacks, thus undermining their efficacy. In this paper, we show that accuracy of deepfake detectors can be considerably improved by incorporating an adversarial learning step during model building. We use a recently proposed deep network architecture, namely VGG19, as deepfake detector supplemented with adversarial training using Iterative Fast Gradient Sign Method (I-FGSM). To further improve non-adversarial accuracy ensemble of models were used. Extensive experiments on a large deepfake video corpus under different white box adversarial attacks demonstrate significant adversarial robustness of the proposed method.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“This technique has been demonstrated to be effective across both visual and audio modalities [202, 203].”
From Deepfake detection across image, video, and audio: a comprehensive survey with empirical evaluation of generalization and robustness (2026) · cites this paper as [202]
In the citing paper
“Pure adversarial training [202, 203] cuts WA ASR from 81 . 06% to 37 . 03% and TA ASR from 34 . 17% to 2 . 01%”
From Deepfake detection across image, video, and audio: a comprehensive survey with empirical evaluation of generalization and robustness (2026) · cites this paper as [202]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 2×
“Various pre-trained models have been made open-source like Xception, Inception [15] and ResNet [16].”
MethodVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 4×
“Models selected to be in ensemble were VGG19, Wide ResNet [26] and Vision Transformer [27].”
MethodInfluentialGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472 · cited 2×
“Models selected to be in ensemble were VGG19, Wide ResNet [26] and Vision Transformer [27].”
MethodXception: Deep Learning with Depthwise Separable Convolutions
Chollet, F. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.195 · Cited by 20,380 · cited 1×
“Likewise, XceptionNet [14] is another CNN architecture that has been shown to achieve high accuracy for the deepfake detection task.”
MethodLabaca-Castro, R. (2023) · Machine Learning under Malware Attack · doi:10.1007/978-3-658-40442-0_9 · Cited by 19,272
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381 · cited 2×
“The inner maximization expression tries to modify the input X such that the loss Adversarial attacks like Fast Gradient Sign Method (FGSM) [23] can be used for generating adversarial data in a single step.”
BackgroundMethodIntriguing properties of neural networks
Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013) · arXiv · doi:10.48550/arxiv.1312.6199 · Cited by 8,143 · cited 1×
“Some of the recent research highlight that deepfake detectors are quite vulnerable to adversarial attacks [18].”
BackgroundZagoruyko, S., & Komodakis, N. (2016) · Procedings of the British Machine Vision Conference 2016 · doi:10.5244/c.30.87 · Cited by 7,244
Towards Deep Learning Models Resistant to Adversarial Attacks
Madry, A., Makelov, A., Schmidt, L., Tsipras, D., & Vladu, A. (2017) · arXiv · doi:10.48550/arxiv.1706.06083 · Cited by 6,602 · cited 1×
“In this section, we present the results of our experiments on three different types of adversarial attacks, namely, Iterative FGSM (I-FGSM) [24], Projected Gradient Descent (PGD) [30] and Momentum Iterative FGSM (MI-FGSM) [31].”
Background
Research works printed but not linked: 1
- Koopman, 2018, Detection of deepfake video manipulation
Cited by
Abbas, F., & Taeihagh, A. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124260 · Cited by 184
“Detecting deepfakes is more challenging than detecting traditional manipulation of digital media due to the minimal divergence between the real and counterfeit data boundaries (Devasthale & Sural, 2022; Kawa & Syga, 2021).”
BackgroundDeepFake on Face and Expression Swap: A Review
Waseem, S., Abu Bakar, S. A. R. S., Ahmed, B. A., Omar, Z., Eisa, T. A. E., & Dalam, M. E. E. (2023) · IEEE Access · doi:10.1109/access.2023.3324403 · Cited by 65
“Developers of deepfake videos are becoming adept at evading detection through the use of advanced techniques such as adversarial generation algorithms, which mask distinct markers or conceal unique identifiers of deepfakes [45].”
BackgroundNguyen-Le, H.-H., Tran, V.-T., Nguyen, D.-T., & Le-Khac, N.-A. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11608-4 · Cited by 28
SLM-DFS: A systematic literature map of deepfake spread on social media
Atlam, E., Almaliki, M., Elmarhomy, G., Almars, A. M., Elsiddieg, A. M. A., & El-Agamy, R. (2024) · Alexandria Engineering Journal · doi:10.1016/j.aej.2024.10.076 · Cited by 26
D4: Detection of Adversarial Diffusion Deepfakes Using Disjoint Ensembles
Hooda, A., Mangaokar, N., Feng, R., Fawaz, K., Jha, S., & Prakash, A. (2024) · 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv57701.2024.00377 · Cited by 7
“Our third baseline ARDD-AT is also from prior work [53], which changes the architectures of ARDD to VGG19, Vision Transformer, Wide-ResNet, and also adversarially trains each individual model using the PGD-150 attack on deepfake images only.”
BackgroundMethod
Metadata
- Authors
- Devasthale, A., & Sural, S.
- Affiliations
- Indian Institute of Technology Kharagpur (IN)
- ORCID
- Sural, Shamik
- Venue
- 2022 IEEE Symposium Series on Computational Intelligence (SSCI)
- Publisher
- IEEE
- Published
- 2022-12-04
- Pages
- 396-403
- OpenAlex
- W4318603549
- Cited by
- 7
- References
- 29
- Citations by year
- 2026: 2 · 2024: 4 · 2023: 1
- FWCI
- 0.45
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Symposium Series on Computational Intelligence (SSCI) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Keywords
- deepfake detection, adversarial robustness, adversarial training, FGSM, VGG19, ensemble learning, fake video detection, white-box attacks
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Adversarially Robust Deepfake Video Detection. https://app.cochat.ai/papers/10.1109/ssci51031.2022.10022079
- The paper
- Devasthale, A., & Sural, S. (2022). Adversarially Robust Deepfake Video Detection. 2022 IEEE Symposium Series on Computational Intelligence (SSCI). https://doi.org/10.1109/ssci51031.2022.10022079
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.