Spatiotemporal Inconsistency Learning for DeepFake Video Detection
Open accessThe DOI matches its Crossref record; 40 of the 40 references it cites were found in a registry or an index.
Abstract
The rapid development of facial manipulation techniques has aroused public concerns in recent years. Following the success of deep learning, existing methods always formulate DeepFake video detection as a binary classification problem and develop frame-based and video-based solutions. However, little attention has been paid to capturing the spatial-temporal inconsistency in forged videos. To address this issue, we term this task as a Spatial-Temporal Inconsistency Learning (STIL) process and instantiate it into a novel STIL block, which consists of a Spatial Inconsistency Module (SIM), a Temporal Inconsistency Module (TIM), and an Information Supplement Module (ISM). Specifically, we present a novel temporal modeling paradigm in TIM by exploiting the temporal difference over adjacent frames along with both horizontal and vertical directions. And the ISM simultaneously utilizes the spatial information from SIM and temporal information from TIM to establish a more comprehensive spatial-temporal representation. Moreover, our STIL block is flexible and could be plugged into existing 2D CNNs. Extensive experiments and visualizations are presented to demonstrate the effectiveness of our method against the state-of-the-art competitors.
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
“However, these two-branch approaches may fragment spatiotemporal cooperation and lead to subtle artifacts being neglected.”
From TALL: Thumbnail Layout for Deepfake Video Detection (2023) · cites this paper as [15]
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset
Carreira, J., & Zisserman, A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.502 · Cited by 9,570
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01228-7 · Cited by 6,909
FaceForensics++: Learning to Detect Manipulated Facial Images
Rossler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., & Niessner, M. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00009 · Cited by 3,314 · cited 4×
“FaceForensics++ (FF++) [26]: FF++ is a standardized benchmark for evaluation of face forgery detection methods.”
BackgroundMethodInfluentialRich Models for Steganalysis of Digital Images
Fridrich, J., & Kodovsky, J. (2012) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2012.2190402 · Cited by 2,144
Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics
Li, Y., Yang, X., Sun, P., Qi, H., & Lyu, S. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00327 · Cited by 1,846 · cited 1×
“Celeb-DF [15]: Celeb-DF is a large-scale dataset in DeepFake detection for developing and evaluating DeepFake detection algorithms, which comprises 5,639 high-quality DeepFake videos from YouTube and contains more than 2 million frames in total.”
MethodDeferred neural rendering: image synthesis using neural textures
Thies, J., Zollhöfer, M., & Nießner, M. (2019) · ACM Transactions on Graphics · doi:10.1145/3306346.3323035 · Cited by 1,356
Face X-Ray for More General Face Forgery Detection
Li, L., Bao, J., Zhang, T., Yang, H., Chen, D., Wen, F., & Guo, B. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00505 · Cited by 1,211
Research works printed but not linked: 3
- Lingzhi Li , Jianmin Bao , Hao Yang , Dong Chen , and Fang Wen . 2019. FaceShifter: Towards High Fidelity And Occlusion Aware Face Swapping. arXiv preprint arXiv:1912.13457 ( 2019 ). Lingzhi Li, Jianmin Bao, Hao Yang, Dong Chen, and Fang Wen. 2019. FaceShifter: Towards High Fidelity And Occlusion Aware Face Swapping. arXiv preprint arXiv:1912.13457 (2019). · arXiv
- Ji Lin , Chuang Gan , and Song Han . 2019 . TSM: Temporal Shift Module for Efficient Video Understanding. In 2019 IEEE/CVF International Conference on Computer Vision. IEEE, 7082--7092 . Ji Lin, Chuang Gan, and Song Han. 2019. TSM: Temporal Shift Module for Efficient Video Understanding. In 2019 IEEE/CVF International Conference on Computer Vision. IEEE, 7082--7092.
- Limin Wang , Zhan Tong , Bin Ji , and Gangshan Wu . 2020 a. TDN: Temporal Difference Networks for Efficient Action Recognition. CoRR , Vol. abs/ 2012 .10071 ( 2020 ). Limin Wang, Zhan Tong, Bin Ji, and Gangshan Wu. 2020 a. TDN: Temporal Difference Networks for Efficient Action Recognition. CoRR, Vol. abs/2012.10071 (2020). · arXiv
Cited by
End-to-End Reconstruction-Classification Learning for Face Forgery Detection
Cao, J., Ma, C., Yao, T., Chen, S., Ding, S., & Yang, X. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.00408 · Cited by 428
“Many efforts have been made to improve the performance of face forgery detection [1,13,23, 29,30,40,45,46,60].”
BackgroundSelf-citationDual Contrastive Learning for General Face Forgery Detection
Sun, K., Yao, T., Chen, S., Ding, S., Li, J., & Ji, R. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i2.20130 · Cited by 209
“Over the past few years, face forgery methods have achieved significant success and received lots of attention in the computer vision community (Thies et al. 2015; Rossler et al. 2019; Dolhansky et al. 2020; Wang et al. 2020; Gu et al. 2021).”
BackgroundSelf-citationAVoiD-DF: Audio-Visual Joint Learning for Detecting Deepfake
Yang, W., Zhou, X., Chen, Z., Guo, B., Ba, Z., Xia, Z., Cao, X., & Ren, K. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3262148 · Cited by 198
“3) Effectiveness of TSE: Recently, some works have exploited spatial-temporal inconsistency for a more general deepfake detection such as STIL [5].”
BackgroundISTVT: Interpretable Spatial-Temporal Video Transformer for Deepfake Detection
Zhao, C., Wang, C., Hu, G., Chen, H., Liu, C., & Tang, J. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3239223 · Cited by 184
“Fortunately, a lot of Deepfake detectors [3], [4], [5], [6], [7], [8] have been devised and they have achieved success on the large-scale Deepfake datasets [9], [10], [11].”
BackgroundMethodInfluentialLeveraging Real Talking Faces via Self-Supervision for Robust Forgery Detection
Haliassos, A., Mira, R., Petridis, S., & Pantic, M. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01453 · Cited by 180
“This technology’s potential social harm has spurred considerable research efforts to detect forgery content [3, 22, 33, 41, 46, 47, 49, 58, 63, 76, 87, 106, 110, 112].”
BackgroundF 2 Trans: High-Frequency Fine-Grained Transformer for Face Forgery Detection
Miao, C., Tan, Z., Chu, Q., Liu, H., Hu, H., & Yu, N. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3233774 · Cited by 172
Masked Relation Learning for DeepFake Detection
Yang, Z., Liang, J., Xu, Y., Zhang, X.-Y., & He, R. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3249566 · Cited by 143
AltFreezing for More General Video Face Forgery Detection
Wang, Z., Bao, J., Zhou, W., Wang, W., & Li, H. (2023) · 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52729.2023.00402 · Cited by 137
“There are A part of works [19, 21, 54] which tend to focus on representation learning.”
BackgroundInfluentialDelving Deeper into Anti-Aliasing in ConvNets
Zou, X., Xiao, F., Yu, Z., Li, Y., & Lee, Y. J. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01672-y · Cited by 138
TALL: Thumbnail Layout for Deepfake Video Detection
Xu, Y., Liang, J., Jia, G., Yang, Z., Zhang, Y., & He, R. (2023) · 2023 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.02071 · Cited by 109
Cai, Z., Stefanov, K., Dhall, A., & Hayat, M. (2022) · 2022 International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta56598.2022.10034605 · Cited by 99
“Previous methods [35], [36] also aim to detect temporal inconsistencies in deepfake content and recently, several audio-visual deepfake detection methods such as MDS [28] and M2TR [37] were proposed.”
MethodExploring Frequency Adversarial Attacks for Face Forgery Detection
Jia, S., Ma, C., Yao, T., Yin, B., Ding, S., & Yang, X. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.00407 · Cited by 97
Self-citationDelving into the Local: Dynamic Inconsistency Learning for DeepFake Video Detection
Gu, Z., Chen, Y., Yao, T., Ding, S., Li, J., & Ma, L. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i1.19955 · Cited by 93
“More recent works start to study the inconsistency to locate the forgery trace in DeepFake videos, e.g., S-MIL (Li et al. 2020a) and STIL (Gu et al. 2021), and show promising results.”
BackgroundSelf-citationDynamic Difference Learning With Spatio–Temporal Correlation for Deepfake Video Detection
Yin, Q., Lu, W., Li, B., & Huang, J. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3290752 · Cited by 92
“Furthermore, STIL [35], Intra-SIM [36], and HCIL [48] captured region-aware inconsistency from local patches and fused snippet-aware features among global snippets.”
BackgroundChen, H., Lin, Y., Li, B., & Tan, S. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3209336 · Cited by 80
“On the other hand, several works [29], [30],”
BackgroundA literature review and perspectives in deepfakes: generation, detection, and applications
Dagar, D., & Vishwakarma, D. K. (2022) · International Journal of Multimedia Information Retrieval · doi:10.1007/s13735-022-00241-w · Cited by 113
FedForgery: Generalized Face Forgery Detection With Residual Federated Learning
Liu, D., Dang, Z., Peng, C., Zheng, Y., Li, S., Wang, N., & Gao, X. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3293951 · Cited by 96
“[25] designed a spatial temporal inconsistency learning module, which takes advantage of the temporal differences between adjacent frames along horizontal and vertical directions.”
BackgroundDetection of Deepfake Videos Using Long-Distance Attention
Lu, W., Liu, L., Zhang, B., Luo, J., Zhao, X., Zhou, Y., & Huang, J. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3233063 · Cited by 72
“The calculation of the motion residual r t can be formulated as The temporal difference operation is simple and does not introduce any extra parameters but is capable of modeling the temporal inconsistency efficiently [56].”
BackgroundA Contemporary Survey on Deepfake Detection: Datasets, Algorithms, and Challenges
Gong, L. Y., & Li, X. J. (2024) · Electronics · doi:10.3390/electronics13030585 · Cited by 75
“Thus, Gu et al. presented a detection work [40] based on learning fake videos’ spatial and temporal inconsistency.”
BackgroundYan, Z., Zhao, Y., Shen, C., Guo, M., Fu, X., Yao, T., Ding, S., Wu, Y., & Li, Y. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01177 · Cited by 116
Self-citationShape Matters: Deformable Patch Attack
Chen, Z., Li, B., Wu, S., Xu, J., Ding, S., & Zhang, W. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19772-7_31 · Cited by 90
“Despite achieving considerably excellent performance on various computer vision tasks [52,53,66,67,68,16,13,14,17,15,26,25,63,46,59,61,4,58], deep neural networks (DNNs) have been shown to be susceptible and vulnerable to adversarial examples, where an adversary introduces an imperceptible…”
BackgroundSelf-citationAugmented Multi-Scale Spatiotemporal Inconsistency Magnifier for Generalized DeepFake Detection
Yu, Y., Zhao, X., Ni, R., Yang, S., Zhao, Y., & Kot, A. C. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3237322 · Cited by 58
“Moreover, for video-based methods, we compare the meth-ods with GIV [18], [19], [20] and LSLT views [23].”
BackgroundMethodInfluentialPVASS-MDD: Predictive Visual-Audio Alignment Self-Supervision for Multimodal Deepfake Detection
Yu, Y., Liu, X., Ni, R., Yang, S., Zhao, Y., & Kot, A. C. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3309899 · Cited by 63
“Therefore, some recent detectors [15], [16], [17], [19], [20], [25], [26], [27], [28] capture spatiotemporal information clues that are vital for visual deepfake detection.”
BackgroundMethodCai, Z., Ghosh, S., Dhall, A., Gedeon, T., Stefanov, K., & Hayat, M. (2023) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2023.103818 · Cited by 63
MCL: Multimodal Contrastive Learning for Deepfake Detection
Liu, X., Yu, Y., Li, X., & Zhao, Y. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3312738 · Cited by 84
“To better demonstrate the superiority of our method, the results are compared with the state-of-the-art Deepfake detection methods, including spatial based methods [11], [13], [28], temporal based methods [18], [19], [48] and multi-modal based methods [20], [21], [22].”
BackgroundMethodMSVT: Multiple Spatiotemporal Views Transformer for DeepFake Video Detection
Yu, Y., Ni, R., Zhao, Y., Yang, S., Xia, F., Jiang, N., & Zhao, G. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3281448 · Cited by 67
“[33] propose the STIL block to capture temporal inconsistency along two orthogonal directions.”
BackgroundMethodInfluentialAn Information Theoretic Approach for Attention-Driven Face Forgery Detection
Sun, K., Liu, H., Yao, T., Sun, X., Chen, S., Ding, S., & Ji, R. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19781-9_7 · Cited by 79
“…faces and the corresponding real faces with some morphological transformations To tackle this issue, many heuristic methods [11, 23, 32, 45, 7, 17] usually use the prior knowledge or observed clues to learn more discriminative features, which are instrumental in distinguishing real and fake…”
MethodSelf-citationLearning Spatiotemporal Inconsistency via Thumbnail Layout for Face Deepfake Detection
Xu, Y., Liang, J., Sheng, L., & Zhang, X. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02054-2 · Cited by 67
“Additionally, [46,47] have attempted to incorporate temporal information by adding modules to image models.”
A Novel Smart Deepfake Video Detection System
Elpeltagy, M., Ismail, A., Zaki, M. S., & Eldahshan, K. (2023) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2023.0140144 · Cited by 39
“The deepfake visual video detection methods can be categorized into Convolution Neural Network (CNN)-based methods [19, 20, 21, 22], methods that are based on CNN with a temporal network [23, 24, 25, 26, 27], handcrafted feature-based methods [28], and handcrafted feature-based methods with deep networks [29, 30].”
MethodDeepFake-Adapter: Dual-Level Adapter for DeepFake Detection
Shao, R., Wu, T., Nie, L., & Liu, Z. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-024-02274-6 · Cited by 48
Hierarchical Contrastive Inconsistency Learning for Deepfake Video Detection
Gu, Z., Yao, T., Chen, Y., Ding, S., & Ma, L. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19775-8_35 · Cited by 86
“…the superiority of our HCIL, we select several video-based detectors for comparison, including Xception [39], VA-LogReg [34], D-FWA [28], FaceNetLSTM [41], Capsule [35], Co-motion [46], DoubleRNN [33], S-MIL [26], DeepRhythm [37], ADDNet-3d [60], STIL [12] , DIANet [18] and TD-3DCNN [55].”
BackgroundMethodInfluentialSelf-citationGenConViT: Deepfake Video Detection Using Generative Convolutional Vision Transformer
Deressa, D. W., Mareen, H., Lambert, P., Atnafu, S., Akhtar, Z., & Van Wallendael, G. (2025) · Applied Sciences · doi:10.3390/app15126622 · Cited by 46
“The visual features-based detection can be on a frame-per-frame basis [22] or temporal relationship between differences be-tween frames through time [18], [29].”
Spatial-Temporal Frequency Forgery Clue for Video Forgery Detection in VIS and NIR Scenario
Yukai, W. A. N. G., Peng, C., Liu, D., Wang, N., & Gao, X. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3281475 · Cited by 51
“Agarwal et al. (Agarwal et al. 2019) found that the tampering with the face area in the video can cause the pattern of facial expressions and head movements inconsistent with the person’s identity when speaking.”
BackgroundMethodHolisticDFD: Infusing spatiotemporal transformer embeddings for deepfake detection
Raza, M. A., Malik, K. M., & Haq, I. U. (2023) · Information Sciences · doi:10.1016/j.ins.2023.119352 · Cited by 46
Interactive Two-Stream Network Across Modalities for Deepfake Detection
Wu, J., Zhang, B., Li, Z., Pang, G., Teng, Z., & Fan, J. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3269841 · Cited by 46
“For instance, Gu et al. [14] employ 3DCNN to detect abnormalities in facial expressions and body movements and achieve good performance.”
BackgroundMethodInfluentialImproving Generalization of Deepfake Detectors by Imposing Gradient Regularization
Guan, W., Wang, W., Dong, J., & Peng, B. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3396064 · Cited by 52
“Some methods train the detectors with synthetic data [8], [11], [12], [32], [33], [34].”
BackgroundMethodYan, Z., Zhao, Y., Chen, S., Guo, M., Fu, X., Yao, T., Ding, S., & Yuan, L. (2024) · arXiv · doi:10.48550/arxiv.2408.17065 · Cited by 37
Self-citationDiff-ID: An Explainable Identity Difference Quantification Framework for DeepFake Detection
Yu, C., Zhang, X., Duan, Y., Yan, S., Wang, Z., Xiang, Y., Ji, S., & Chen, W. (2024) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2024.3364679 · Cited by 38
“Some approaches detect the irregular jitter [28], [29], [30] of fake videos on the time series.”
BackgroundAdaptive Face Forgery Detection in Cross Domain
Song, L., Fang, Z., Li, X., Dong, X., Jin, Z., Chen, Y., & Lyu, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19830-4_27 · Cited by 49
“Following the previous studies [22, 55, 56], the video-level AUC scores on DFDC and Celeb-DF v2 are presented in experiments.”
MethodSelf-citationContrastive Pseudo Learning for Open-World DeepFake Attribution
Sun, Z., Shen, C., Yao, T., Yin, B., Yi, R., Ding, S., & Ma, L. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01909 · Cited by 57
“A plethora of works [16, 60, 50, 72, 12, 7, 59, 58, 22, 23, 26, 25, 24] for the real/fake detection task have been proposed in recent years.”
BackgroundSelf-citationSoK: Systematization and Benchmarking of Deepfake Detectors in a Unified Framework
Le, B. M., Kim, J., Woo, S. S., Moore, K., Abuadbba, A., & Tariq, S. (2025) · IEEE European Symposium on Security and Privacy · doi:10.1109/eurosp63326.2025.00055 · Cited by 43
MINTIME: Multi-Identity Size-Invariant Video Deepfake Detection
Coccomini, D. A., Zilos, G. K., Amato, G., Caldelli, R., Falchi, F., Papadopoulos, S., & Gennaro, C. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3409054 · Cited by 33
Deep learning technology for face forgery detection: A survey
Ma, L., Yang, P., Xu, Y., Yang, Z., Li, P., & Huang, H. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.129055 · Cited by 41
Self-citationRegion-Aware Temporal Inconsistency Learning for DeepFake Video Detection
Gu, Z., Yao, T., Chen, Y., Yi, R., Ding, S., & Ma, L. (2022) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2022/129 · Cited by 42
Self-citationAn adversarial attack approach for eXplainable AI evaluation on deepfake detection models
Gowrisankar, B., & Thing, V. L. L. (2024) · Computers & Security · doi:10.1016/j.cose.2023.103684 · Cited by 30
“The second reason is the fact that recent deepfake detectors rely on finding spatial inconsistency among frames [35].”
BackgroundTowards Unified Defense for Face Forgery and Spoofing Attacks via Dual Space Reconstruction Learning
Cao, J., Zhang, K.-Y., Yao, T., Ding, S., Yang, X., & Ma, C. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02151-2 · Cited by 35
“Many efforts have been made to improve the performance of face forgery detection (Afchar et al., 2018; Nguyen et al., 2019a; Li et al., 2020b; Masi et al., 2020; Zi et al., 2020; Wang & Deng, 2021; Wang et al., 2020a; Sun et al., 2022; Gu et al., 2021).”
Self-citationNguyen-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
Deepfake Video Detection via Facial Action Dependencies Estimation
Tan, L., Wang, Y., Wang, J., Yang, L., Chen, X., & Guo, Y. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i4.25658 · Cited by 40
“( Gu et al. 2021) gradually amplifies the pixel jitters in the manipulated videos.”
BackgroundHigh-compressed deepfake video detection with contrastive spatiotemporal distillation
Zhu, Y., Zhang, C., Gao, J., Sun, X., Rui, Z., & Zhou, X. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126872 · Cited by 27
DeepFake video detection: Insights into model generalisation — A Systematic review
Ramanaharan, R., Guruge, D. B., & Agbinya, J. I. (2025) · Data and Information Management · doi:10.1016/j.dim.2025.100099 · Cited by 48
WMamba: Wavelet-based Mamba for Face Forgery Detection
Peng, S., Zhang, T., Gao, L., Zhu, X., Zhang, H., Pang, K., & Lei, Z. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755592 · Cited by 32
“In contrast, time-domain methods analyze inter-frame inconsistencies, leveraging the entire video as input to identify manipulations [14– 16, 19, 46, 67, 70].”
Liu, D., Chen, T., Peng, C., Wang, N., Hu, R., & Gao, X. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3516561 · Cited by 29
“…95.41 74.11 - - TEI [41] 96.54 74.66 - - TAM [41] 97.04 67.96 - - V4D [41] 96.74 70.08 - - DSANet [41] 96.88 73.71 - - TD-3DCNN [42] - 57.32 - - DoubleRNN [17] 93.18 73.41 - - ADDNet-3D [32] 96.22 60.85 - - STIL [43] 97.12 75.58 - - HCIL [41] 98.32 79.00 - - Forensic [44] 99 the binary operation.”
BackgroundInfluentialDeepfakes in digital media forensics: Generation, AI-based detection and challenges
Bendiab, G., Haiouni, H., Moulas, I., & Shiaeles, S. N. (2024) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2024.103935 · Cited by 35
Usmani, S., Kumar, S., & Sadhya, D. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.129256 · Cited by 32
DeepMark: A Scalable and Robust Framework for DeepFake Video Detection
Li, T., Ye, Q., Hu, H., Xue, Q., Xiao, Y., & Li, J. (2024) · ACM Transactions on Privacy and Security · doi:10.1145/3629976 · Cited by 26
Self-citationFine-Grained Multimodal DeepFake Classification via Heterogeneous Graphs
Yin, Q., Lu, W., Cao, X., Luo, X., Zhou, Y., & Huang, J. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02128-1 · Cited by 23
“Semantic feature based methods aim to model temporal inconsistency by extracting trends in irregular facial movements (Haliassos et al., 2021, 2022) or differences over adjacent frames (Li et al., 2020b; Gu et al., 2021; Lu et al., 2023).”
Combating deepfakes: a comprehensive multilayer deepfake video detection framework
Rathoure, N., Pateriya, R. K., Bharot, N., & Verma, P. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-20012-5 · Cited by 27
PIM-Net: Progressive Inconsistency Mining Network for image manipulation localization
Bai, N., Wang, X., Han, R., Hou, J., Wang, Y., & Pang, S. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.111136 · Cited by 23
D3: Training-Free AI-Generated Video Detection Using Second-Order Features
Zheng, C., Suo, R., Lin, C., Zhao, Z., Yang, L., Liu, S., Yang, M., Wang, C., & Shen, C. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01194 · Cited by 58
“We report the results of the baselines [16, 19, 24, 34, 50, 58] on GenVideo dataset from [16].”
TCSD: Triple Complementary Streams Detector for Comprehensive Deepfake Detection
Liu, X., Yu, Y., Li, X., Zhao, Y., & Guo, G. (2023) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3558004 · Cited by 27
“Some simple neural network–based methods were proposed early on, including MesoNet [4], XceptionNet [52], and Recurrent Neural Network (RNN) [20, 54, 58].”
BackgroundMethodSaif, S., Tehsin, S., & Ali, S. S. (2024) · Technological Forecasting and Social Change · doi:10.1016/j.techfore.2024.123471 · Cited by 26
Swin-Fake: A Consistency Learning Transformer-Based Deepfake Video Detector
Gong, L. Y., Li, X. J., & Chong, P. H. J. (2024) · Electronics · doi:10.3390/electronics13153045 · Cited by 22
Kundu, R., Xiong, H., Mohanty, V., Balachandran, A., & Roy–Chowdhury, A. K. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.02612 · Cited by 58
Vulnerability-Aware Spatio-Temporal Learning for Generalizable Deepfake Video Detection
Nguyen, D., Astrid, M., Kacem, A., Ghorbel, E., & Aouada, D. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01004 · Cited by 37
“Traditional methods [20, 23, 25, 56, 57, 64, 67, 71] aim to learn jointly a feature extractor Φ : V (cid:55)→ F and a binary classifier f : F → { 0 , 1 } by minimizing the standard binary cross-entropy (BCE) loss L BCE ( f (Φ( X i )) , y i ) using the entire training set V , with F being the…”
Hearing and Seeing Abnormality: Self-Supervised Audio-Visual Mutual Learning for Deepfake Detection
Sung, C.-S., Chen, J.-C., & Chen, C. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095247 · Cited by 29
“Therefore, some studies have focused on model architecture to judge discontinuities between images [10, 11] or differences between audio and visual [12, 13].”
BackgroundRobust Frame-Level Detection for Deepfake Videos With Lightweight Bayesian Inference Weighting
Zhou, L., Ma, C., Wang, Z., Zhang, Y., Shi, X., & Wu, L. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3337128 · Cited by 22
“For instance, STIL [23] designed a spatial–temporal inconsistency learning block to extract the fine-grained temporal features of videos for classifying videos.”
MethodDeepFake detection method based on multi-scale interactive dual-stream network
Cheng, Z., Wang, Y., Wan, Y., & Jiang, C. (2024) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2024.104263 · Cited by 19
Magnifying multimodal forgery clues for Deepfake detection
Liu, X., Yu, Y., Li, X., & Zhao, Y. (2023) · Signal Processing Image Communication · doi:10.1016/j.image.2023.117010 · Cited by 30
Analyzing temporal coherence for deepfake video detection
Amin, M. A., Hu, Y., & Hu, J. (2024) · Electronic Research Archive · doi:10.3934/era.2024119 · Cited by 24
“Gu et al. [10] present the spatial-temporal inconsistency learning (STIL) block to grasp temporal in-coherence along two orthogonal paths.”
DIP: Diffusion Learning of Inconsistency Pattern for General DeepFake Detection
Nie, F., Ni, J., Zhang, J., Zhang, B., & Zhang, W. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521766 · Cited by 30
“Index Terms —Deepfake detection, vision transformer, graph diffusion learning.”
Fighting Malicious Media Data: A Survey on Tampering Detection and Deepfake Detection
Wang, J., Li, Z., Zhang, C., Chen, J., Wu, Z., Davis, L. S., & Jiang, Y. (2025) · Proceedings of the IEEE · doi:10.1109/jproc.2025.3576367 · Cited by 21
“Among them, a considerable amount of work [357], [358], [359] focuses on capturing the spatial– temporal inconsistency of suspicious videos.”
Scalable Universal Adversarial Watermark Defending Against Facial Forgery
Qiao, T., Zhao, B., Shi, R., Han, M., Hassaballah, M., Retraint, F., & Luo, X. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3460387 · Cited by 20
“Some methods [3], [4], [5], [6], [7], [8] refer to training a binary classifier determining whether an inquiry image or video is forged; some methods [9], [10], [11] refer to utilizing the inconsistency of features in forged region to detect fake media.”
MethodSemantics-Oriented Multitask Learning for DeepFake Detection: A Joint Embedding Approach
Zou, M., Yu, B., Zhan, Y., Lyu, S., & Ma, K. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3572508 · Cited by 22
“Gu et al. [67] and Zheng et al. [68] examined spatiotemporal inconsistencies at the signal level, while Haliassos et al. [29] did so at the semantic level through lipreading pretraining.”
Multi-definition Deepfake detection via semantics reduction and cross-domain training
Zhao, C., Wang, C., Song, Z., Hu, G., Wang, L., & Miao, D. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.111469 · Cited by 17
Rethinking prediction alignment in one-stage object detection
Xiao, J., Jiang, H., Li, Z., & Gu, Q. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.09.132 · Cited by 17
Spatiotemporal Inconsistency Learning and Interactive Fusion for Deepfake Video Detection
Zhang, D., Zhu, W., Liao, X., Qi, F., Yang, G., & Ding, X. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3664654 · Cited by 24
“For example, [11, 12] exploited the spatiotemporal inconsistency information of fake videos and obtained better generalization Methods Celeb-DF DFDC Xception [44] 65.50% 59.39% I3D [5] 74.11% 66.87% VA-LogReg [34] 55.10% − TEI [30] 74.66% 67.42% D-FWA [22] 56.90% − Capsule [36] 57.50% − V4D [62]…”
InfluentialRethinking Open-World DeepFake Attribution with Multi-perspective Sensory Learning
Sun, Z., Chen, S., Yao, T., Yi, R., Ding, S., & Ma, L. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02184-7 · Cited by 25
“…numerous works (Chollet, 2017; Tan & Le, 2019; Qian et al., 2020; Zhao et al., 2021; Chen et al., 2022; Cao et al., 2022; Sun et al., 2022b,a; Qiqi et al., 2022; Gu et al., 2021, 2022b,a; Zhihao et al., 2022; Zheng, 2023; Zheng et al., 2024) have been proposed for the real/fake detection task.”
Self-citationTemporal surface frame anomalies for deepfake video detection
Ciamarra, A., Caldelli, R., & Bimbo, A. D. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00388 · Cited by 20
“Gu et al. [6] integrate both spatial and temporal features in a unified 2D CNN framework.”
BackgroundSFormer: An end-to-end spatio-temporal transformer architecture for deepfake detection
Kingra, S., Aggarwal, N., & Kaur, N. (2024) · Forensic Science International Digital Investigation · doi:10.1016/j.fsidi.2024.301817 · Cited by 23
Towards generalizable face forgery detection via mitigating spurious correlation
Bai, N., Wang, X., Han, R., Hou, J., Wang, Q., & Pang, S. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106909 · Cited by 21
Deepfake Face Detection and Adversarial Attack Defense Method Based on Multi-Feature Decision Fusion
Lei, S., Song, J., Feng, F., Yan, Z., & Wang, A. (2025) · Applied Sciences · doi:10.3390/app15126588 · Cited by 21
Multi-level feature disentanglement network for cross-dataset face forgery detection
Fu, Z., Chen, X., Liu, D., Qu, X., Dong, J., Zhang, X., & Ji, S. (2023) · Image and Vision Computing · doi:10.1016/j.imavis.2023.104686 · Cited by 15
Deep Learning Methods to Detect Image Falsification
Janutėnas, L., Janutėnaitė-Bogdanienė, J., & Šešok, D. (2023) · Applied Sciences · doi:10.3390/app13137694 · Cited by 14
Spatio-Temporal Catcher: A Self-Supervised Transformer for Deepfake Video Detection
Li, M., Li, X., Yu, K., Deng, C. D., Huang, H., Mao, F., Xue, H., & Li, M. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3613842 · Cited by 23
Constructing Spatio-Temporal Graphs for Face Forgery Detection
Shang, Z., Xie, H., Yu, L., Zha, Z.-J., & Zhang, Y. (2023) · ACM Transactions on the Web · doi:10.1145/3580512 · Cited by 15
“Gu et al. [23] present a temporal module to detect the temporal incoherence of adjacent frames along with horizontal and vertical directions, respectively.”
BackgroundMethodInfluentialTemporal Diversified Self-Contrastive Learning for Generalized Face Forgery Detection
Zhang, R., He, P., Li, H., Wang, S., & Cao, Y. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3436554 · Cited by 13
“Therefore, recent works adopt video-based detection methods [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], which capture temporal incoherence introduced by the forgery process for detection.”
FAME: a lightweight spatio-temporal network for model attribution of face-swap deepfakes
Ahmad, W., Peng, Y., & Chang, Y.-H. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.128571 · Cited by 14
“Learning-based methods leverage deep neural networks, including capsule networks [43], ensemble CNNs [9], attention models [55], and spatio-temporal architectures [23, 32].”
Local artifacts amplification for deepfakes augmentation
Peng, C., Sun, F., Liu, D., Wang, N., & Gao, X. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106692 · Cited by 12
Ahmad, W., Peng, Y., Chang, Y.-H., Ganfure, G. O., & Khan, S. (2025) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3715138 · Cited by 12
“Interestingly, a multitude of deep-fake detectors, including [23, 35, 72], have been formulated and demonstrated effectiveness in expansive deep-fake datasets [16, 39, 51].”
Localization and detection of deepfake videos based on self-blending method
Xu, J., Liu, X., Lin, W., Shang, W., & Wang, Y. (2025) · Scientific Reports · doi:10.1038/s41598-025-88523-1 · Cited by 16
Zhang, G., Liang, Y., Tian, K., Yi, J., Alsolai, H., Liu, M., & Hu, X. (2025) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2025.3624764 · Cited by 14
“…LRL [36], MADD [37], RECCE [10], MC-LCR [38], M2TR [8], and BRCNet [39]; along with methods focusing on spatiotemporal information mining: STIL [16], Delv-ing [40], TSCL [15], MRE-Net [43], SIA [44], RECCE [10], Intra-SIM [28], MRL [45], CDIN [17], Dynamic [14], Add-Net [35], MultiAtt [37],…”
Illumination Enlightened Spatial-temporal Inconsistency for Deepfake Video Detection
Tian, K., Chen, C., Zhou, Y., & Hu, X. (2024) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme57554.2024.10687905 · Cited by 18
“Celeb-DF DFDC SIA [17] 99.96 90.96 RECCE [18] 99.94 - M2TR [20] 99.80 - Lips [21] - 73.50 STIL [22] 99.78 89.80 Intra-SIM [23] 99.61 92.79 HCIL [24] 99.81 95.11 MRL [25] 99.96 99.11 MRE-Net [26] - 99.75 CDIN [27] 99.70 - ours 99.77 99.94”
InfluentialFine-grained deepfake detection based on cross-modality attention
Zhao, L., Zhang, M., Ding, H., & Cui, X. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08271-z · Cited by 11
“In the meanwhile, spatiotemporal inconsistency of forgery videos has also been noticed in the following works [7].”
BackgroundInfluentialMulti-Scale High-Frequency Vision Transformer for Face Forgery Detection
Chen, Z., Li, H., & Hu, J. (2025) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2025.3543692 · Cited by 11
“Researchers have made a lot of achievements in the field of face forgery detection, and they also put forward a variety of problems and solutions [1], [3], [23], [24], [25], [26], [27], [28], [29].”
Multi-modal Deepfake Detection via Multi-task Audio-Visual Prompt Learning
Miao, H., Guo, Y., Liu, Z., & Wang, Y. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i1.32042 · Cited by 20
“To address this limitation, a series of meth-ods (Sabir et al. 2019; Gu et al. 2021, 2022) have been proposed to adaptively learn the spatial and temporal features, by constructing various network architecturess (Zheng et al. 2021; Xu et al. 2023, 2024), utilizing different learning strategies…”
FauForensics: Boosting Audio-Visual Deepfake Detection With Facial Action Units
Wang, J., Wu, B., Liu, L., & Liu, Q. (2026) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2026.3674401 · Cited by 14
“Previously, some work [38], [39], [40], [41], [42] attempted to capture temporal forgery cues for unimodal deepfake detection.”
Cross-Domain Local Characteristic Enhanced Deepfake Video Detection
Liu, Z., Wang, H., & Wang, S. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-26348-4_12 · Cited by 12
Semantic Token Transformer for Face Forgery Detection
Peng, C., Luo, X., Liu, D., Wang, N., Hu, R., & Gao, X. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3567110 · Cited by 10
Vision Transformer-Based Video Hashing Retrieval for Tracing the Source of Fake Videos
Pei, P., Zhao, X., Li, J., Cao, Y., & Lai, X. (2023) · Security and Communication Networks · doi:10.1155/2023/5349392 · Cited by 11
Self-citationMining Generalized Multi-timescale Inconsistency for Detecting Deepfake Videos
Yu, Y., Ni, R., Yang, S., Ni, Y., Zhao, Y., & Kot, A. C. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-024-02249-7 · Cited by 7
“Tocapturesubtleanomalies,severaldetectorshave concentrated on local spatial regions to pinpoint minuscule defects left behind by different forgery methods (Chai et al., 2020; Zhao et al., 2021; Gu et al., 2022; Chen et al., 2021; Liu et al., 2021; Zhao et al., 2021).”
Influential
Metadata
- Authors
- Gu, Z., Chen, Y., Yao, T., Ding, S., Li, J., Huang, F., & Ma, L.
- Affiliations
- Shanghai Jiao Tong University (CN)
- Tencent (China) (CN)
- Venue
- Proceedings of the 29th ACM International Conference on Multimedia
- Publisher
- ACM
- Published
- 2021-10-17
- Pages
- 3473-3481
- OpenAlex
- W3207531490
- Cited by
- 189
- References
- 36
- Citations by year
- 2026: 36 · 2025: 53 · 2024: 37 · 2023: 44 · 2022: 19
- FWCI
- 8.18
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- ACM International Conference on Multimedia (ACM MM) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Key Research and Development Program of China (No. 2019YFC1521104)
- National Natural Science Foundation of China (61972157)
- Shanghai Municipal Science and Technology Major Project (2021SHZDZX0102)
- Zhejiang Lab (2020NB0AB01)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Video Quality Assessment
- Keywords
- deepfake detection, video forgery detection, face manipulation, spatial inconsistency, time inconsistency, temporal modeling, 2D convolutional neural network
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.