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
Items 201–209 of 209
Mind the Rift: Cross-Scale Coupling Mismatch for AI-Generated Video Detection
Li, S., Yang, J., & Liang, W. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3835693
Comprehensive Survey with Insights from the WildDeepfake Benchmark
Mohsen, M. M., & Obaid, A. J. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-27708-4_25
Breaking Rigidity in Adversarial Patch Attacks
Kumar, V., Balakrishnan, G., & Agarwal, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37369-4_4
A Two-Tiered Framework Based on Deep Learning and Blockchain for Detection of Deepfake Videos
Mishra, A., Saroj, S. K., & Tiwari, R. K. (2026) · International Conference on Green Technologies and Sustainable Solutions (GTSS) · doi:10.1109/gtss68743.2026.11679681
An Exploration of Various Image and Video Deepfake Detection Methods and Models
Kaware, M. A. S. (2025) · Vidhyayana · doi:10.58213/pygxh917
Slat, C. C., & Nurtanio, I. (2026) · IEEE International Conference on Artificial Intelligence and Mechatronics Systems (AIMS) · doi:10.1109/aims71270.2026.11712590
“As demonstrated in spatiotemporal baselines [10], authentic camera sequences naturally present localized, noisier optical flow patterns around dynamic regions, whereas manipulated counterparts display structurally smoother fields.”
RFEA: Reliable Forensic Evidence Aggregation for Deepfake Video Detection
Lu, S., Zhou, Y., Chang, Z., Hao, Q., & Meng, X. (2026) · Applied Sciences · doi:10.3390/app162010170
Evaluating Ensemble Learning for Robust Real Time Deepfake Detection
Gil, M. J., Folajimi, Y., Othman, S., & Deligiannidis, L. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-28469-3_8
MDD-CR: End-to-end multimodal deepfake detection fusing spatio-temporal and spectrographic features
Das, A. K., & Naskar, R. (2026) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2026.105021
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.