Automated face swapping and its detection
The DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.
Abstract
Face has been used as one of the mainstream manners for user identification. However, with the popularity of face-swapping apps, it takes only a few seconds to change the faces between two facial images. Such swapped results, when using improperly or carelessly, might create some security issues in certain applications. This paper is the first work to address the importance of this issue and discusses the feasibility to achieve an automated face swapping detection through machine learning. Several approaches are tested on a face swapping database derived from a face benchmarking repository. The best solution in the experiments achieved a detection accuracy of over 92%.
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
“Recent works on adversarial perturbation attacks to fool DNN-based detectors make the deepfake detection task more difficult [157, 171–174].”
From Deep learning for deepfakes creation and detection: A survey (2022) · cites this paper as [174]
Citations
Cites
LIBSVM: A library for support vector machines
Chang, C.-C., & Lin, C.-J. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1961189.1961199 · Cited by 40,918 · cited 12×
“Compared to current popular deep learning tools that require large dateset of hundred-thousands of images for training [20], SVM [17] is a lite tool that can be easily applied to the relatively small-scale task in this paper.”
BackgroundMethodInfluentialLabeled Faces in the Wild: A Database forStudying Face Recognition in Unconstrained Environments
Huang, G. B., Mattar, M., Berg, T. L., & Miller, E. G. (2008) · HAL · Cited by 4,822
Learning Face Representation from Scratch
Yi, D., Lei, Z., Liao, S., & Li, S. Z. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1411.7923 · Cited by 1,521 · cited 1×
“Compared to current popular deep learning tools that require large dateset of hundred-thousands of images for training [20], SVM [17] is a lite tool that can be easily applied to the relatively small-scale task in this paper.”
MethodSplines minimizing rotation-invariant semi-norms in Sobolev spaces
Duchon, J. (1977) · Lecture Notes in Mathematics · doi:10.1007/bfb0086566 · Cited by 1,380 · cited 1×
“2) Align two faces using the detected landmarks based on the thin-plate spline model [14].”
MethodEnsemble Classifiers for Steganalysis of Digital Media
Kodovsky, J., Fridrich, J., & Holub, V. (2012) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2011.2175919 · Cited by 1,064 · cited 3×
“The RF takes the default setting in [18] and the hidden layer size of the MLP is set to 800.”
BackgroundMethodFace Spoof Detection With Image Distortion Analysis
Wen, D., Han, H., & Jain, A. K. (2015) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2015.2400395 · Cited by 905 · cited 1×
“Specifically, current falsifying data are fake copies of an authorized client's face, in forms such as photos or videos [5, 6].”
Background(2012) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · Cited by 826
A face antispoofing database with diverse attacks
Zhang, Z., Yan, J., Liu, S., Lei, Z., Yi, D., & Li, S. Z. (2012) · 2012 5th IAPR International Conference on Biometrics (ICB) · doi:10.1109/icb.2012.6199754 · Cited by 816 · cited 1×
“Specifically, current falsifying data are fake copies of an authorized client's face, in forms such as photos or videos [5, 6].”
BackgroundDiscriminant Analysis of Principal Components for Face Recognition
Zhao, W., Krishnaswamy, A., Chellappa, R., Swets, D. L., & Weng, J. (1998) · Face Recognition · doi:10.1007/978-3-642-72201-1_4 · Cited by 696
LBP − TOP Based Countermeasure against Face Spoofing Attacks
de Freitas Pereira, T., Anjos, A., De Martino, J. M., & Marcel, S. (2013) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-37410-4_11 · Cited by 319
Research works printed but not linked: 7
- 0, Bian Lian
- 0, Face Swap Booth-photo faceswap & face changer
- 0, Auto Face Swap
- 0, Face Swap Live
- bay, 2006, Surf: Speeded up robust features[C], European Conference on Computer Vision, 404
- wen, 2015, Face spoof detection with image distortion analysis, Information Forensics and Security, IEEE Transactions on, 10, 746
- 0, Windows Hello face authentication
Cited by
Items 101–155 of 155
Deep Learning based Model for Deepfake Image Detection: An Analytical Approach
Neha, & Arora, B. (2023) · International Conference on Innovative Mechanisms for Industry Applications · doi:10.1109/icimia60377.2023.10426561 · Cited by 6
“To differentiate real and face swap images, [36] employed the bag of words method to extract a set of compact features that were then fed into a variety of classifiers, including SVM, RF, and MLP.”
MethodExposing the Forgery Clues of DeepFakes via Exploring the Inconsistent Expression Cues
Liu, J., Wang, L., Wang, R., Ke, J., Ye, X., & Wu, Y. (2025) · International Journal of Intelligent Systems · doi:10.1155/int/7945646 · Cited by 7
“Te identity swap can replace the face between the target and source image, and CycleGAN [20] and FaceSwap [21] are the two classical works.”
Meel, P., Raj, C., & Bhawna (2025) · International Journal of Data Science and Analytics · doi:10.1007/s41060-025-00727-w · Cited by 9
“[92] Fake news detection Image CNN, Iterative Transfer Learning Binary Weibo 77%”
Faceswap Finder: A Fusion-Based Deepfake Detection Technique
Basit, N., Khalid, F., Ain, Q. U., & Siddiqui, M. A. (2025) · International Conference on Advancements in Computational Sciences (ICACS) · doi:10.1109/icacs64902.2025.10937811 · Cited by 5
“The SURF descriptor and the support vector machine (SVM) classifier were utilized in [20] to detect faceswaps, yet this method was unable to identify videos that had been altered or manipulated.”
Face Swapping for Solving Collateral Privacy Issues in Multimedia Analytics
Werner Bailer, A. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-05710-7_14 · Cited by 5
A dual-stream model based on PRNU and quaternion RGB for detecting fake faces
Hua, M., Li, S., & Wang, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0314041 · Cited by 4
Lin, Y., Qu, Y., Li, Y., & Nie, Z. (2020) · International Conference on Computational Science and Computational Intelligence (CSCI) · doi:10.1109/csci51800.2020.00291 · Cited by 3
SE-ResNet56: Robust Network Model for Deepfake Detection
Wang, X., Zhao, Z., Zhang, C., Bai, N., & Hu, X. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-25115-3_3 · Cited by 5
Local similarity anomaly for general face forgery detection
Dai, Y., Fei, J., Xia, Z., Liu, J.-N., & Weng, J. (2023) · Journal of Image and Graphics · doi:10.11834/jig.221006 · Cited by 3
Convolutional Neural Network for Detecting Deepfake Palmprint Images
Tsai, M.-J., & Cheng-Tao, C. (2024) · IEEE Access · doi:10.1109/access.2024.3433497 · Cited by 7
“Zhang et al. [8] utilized the Speeded Up Robust Features (SURF) method to pinpoint critical elements for a solution.”
MethodDigital Forensics Triage Classification Model using Hybrid Learning Approaches
Mohmed Afridhi, L., & Palanivel, K. (2022) · International Journal of Innovative Research in Computer Science & Technology · doi:10.55524/ijircst.2022.10.3.7 · Cited by 4
The Defensive Methods Against Deepfake: Review
Talib, E., Hassan, N. F., & Jamil, A. S. (2023) · Iraqi Journal of Science · doi:10.24996/ijs.2023.64.10.39 · Cited by 7
Hybrid CNN-based detection of forged facial images using Gabor filters and local binary patterns
Kaman, S., Javeriya, S. B., & Makandar, A. (2025) · Discover Computing · doi:10.1007/s10791-025-09731-x · Cited by 5
Ensemble Learning Model for Face Swap Detection
Samrouth, K., Beuve, N., Déforges, O., Bakir, N., & Hamidouche, W. (2024) · International Symposium on Digital Forensics and Security (ISDFS) · doi:10.1109/isdfs60797.2024.10527286 · Cited by 6
“On one hand, few solutions were based on handcrafted features extracted using common machine vision tools such as SURF (Speeded Up Robust Features) [9] and multimedia stream descriptors [10].”
MethodReal Time Face Swap Based on Patch Warping
Şenses, M., & Topal, C. (2019) · Signal Processing and Communications Applications Conference (SIU) · doi:10.1109/siu.2019.8806397 · Cited by 2
“Bunlardan bazıları yüz değişimi yapılmış yüzlerin tespiti [2], kullanıcı gizliliğini korumak, dijital adli tıp oluşturulması [4] ve potansiyel yüze ait veri arttırımı ile duygu tanıma [5], [6] olarak sıralanabilir.”
Identification of Deepfakes using Strategic Models and Architectures
Nallapati, S. R., Dommeti, D., Medhalavalasa, S., Bonku, K. K., Srinivas, P. V. V. S., & Bhattacharyya, D. (2023) · International Conference on Sustainable Computing and Data Communication Systems (ICSCDS) · doi:10.1109/icscds56580.2023.10104880 · Cited by 2
“[11] discuss the benefits and drawbacks of existing works.”
BackgroundExploring Deepfakes and Effective Prevention Strategies: A Critical Review
Garcia, J. M. S. (2025) · Psychology and Education A Multidisciplinary Journal · doi:10.70838/pemj.330107 · Cited by 1
Deepfake Detection Using Deep Learning and Evolutionary Optimization
Gaur, A., Jain, B., Kushwaha, A., Dahiya, A., & Saigal, S. (2025) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-96-0451-7_16 · Cited by 1
Media Forensics and Deepfake-Systematic Survey
Nadeem Jabbar, C. H., Saghir, A., Meer, A. A., Sahi, S., Hassan, B., & Yasir, S. M. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-77617-5_24 · Cited by 6
Zou, Z., Peng, D., Zhao, Y., Tian, Z., & Cai, J. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06677-x · Cited by 2
“Zhang et al. [ 19] were the first to use classical machine learning methods to tackle the issue of face-swapped image detection, such as support vector machines, random forest, and multilayer perceptron.”
Learnable Gabor and Wavelet Filters Based Deep Learning Method for Image Forgery Detection
Joshi, P., Muthu, M. J. S., & Masilamani, V. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-93709-5_28 · Cited by 2
Diffusion-Aware Deepfake Detection via Kernel PCA on Noise Residuals for Semantic Communication
Chen, B. (2025) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2025-spring65109.2025.11174885 · Cited by 1
“Simularly, Zhang et al.[15] employs bag-of-words features with traditional classifiers such as SVMs [16], random forests [17], and MLPs [18].”
Content-Aware Face Blending by Label Propagation
Liang, L., & Zhang, X. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-03338-5_15 · Cited by 1
Recent Trends in Deepfake Detection
Sudarshana, K., & Mylarareddy, C. (2021) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-7998-7728-8.ch001 · Cited by 2
Chithra Raj, N., Dutta, M., & Saini, J. (2024) · International Conference on Trends in Engineering Systems and Technologies (ICTEST) · doi:10.1109/ictest60614.2024.10576181 · Cited by 1
“1: Swapping two faces using the “autofaceswap” software (of picture courtesy [9]) Image splicing manipulation combines sections from different images, creating a new, often deceptive context, as shown in Fig.”
MethodResearch on Video Tampering Detection Method Based on Multilevel Smoothness Analysis
Chen, Z., & Cheng, X. (2024) · IEEE Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) · doi:10.1109/itnec60942.2024.10733140 · Cited by 1
“For example, Zhang et al. [1] used manually designed rules to distinguish forged data and achieved good classification results with simple training.”
BackgroundExploring Feature Map Correlations for Effective Fake Face Detection
Honarjoo, N., Taher, F., & Mansouri, A. (2024) · International Symposium on Telecommunications · doi:10.1109/ist64061.2024.10843660 · Cited by 2
“[18] emphasized changing feces between tivo facial images and proposed a mefood for automated fece-swappfog detection forough machine learning.”
Combating Deepfake-Generated Photos and Videos Using Generative Adversarial Network
Aarthi, B., Smruthi, A., Thanishka, P., Prasanna, G. S., & Mahendran, P. K. R. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-4672-4.ch003 · Cited by 1
Vis‐a‐Vis: A Tool for Face Components Replacement
Lestriandoko, N. H., Spreeuwers, L., & Veldhuis, R. N. J. (2025) · IET Image Processing · doi:10.1049/ipr2.70212
Deep Residual Learning based Discriminator for Identifying Deepfakes with Cut-Out Regularization
Gummadi, S. D., & Ghosh, A. (2022) · IEEE World Conference on Applied Intelligence and Computing (AIC) · doi:10.1109/aic55036.2022.9848964 · Cited by 2
“In [15] the authors used the bag of words technique to convert images into histograms of descriptors belonging to each word and classical machine learning classifier models (SVM and Random Forest) to train on various features.”
MethodDeepfake Detection with Data Privacy Protection
Wu, M., Wang, F., Wu, X., Yu, F., Wang, B., & Song, Z. (2022) · 2022 IEEE 24th International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp55362.2022.9949458
“For example, people propose features like Speeded-Up Robust Features(SURF) [12], Photo Response Non-Uniformity(PRNU) [13], and use machine learning classifiers to obtain predictors (classifiers).”
MethodAutomated Face Detection And Swapping In Video
Patel, J., Shetty, A., Vishwakarma, A., & Shelke, V. (2021) · International Conference on Advanced Scientific Innovation in Science, Engineering and Technology (ICASISET) · doi:10.4108/eai.16-5-2020.2303961
“Existing System Ying Zhang [1] Shape Predictor 86 Facial Landmark s”
BackgroundMethodInfluentialChallenges and Solutions in DeepFakesPreprint
Sharma, J., & Sharma, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.05397 · Cited by 2
Raj, C., & Meel, P. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2112.00803
“[92] Fake news detection Image CNN, Iterative Transfer Learning Binary Weibo 77%”
Довгаль, В. А. (2022) · Вестник Адыгейского государственного университета серия «Естественно-математические и технические науки» · doi:10.53598/2410-3225-2021-4-291-82-94 · Cited by 4
Contrastive Knowledge Transfer for Deepfake Detection with Limited Data
Li, D., Zhuo, W., Wang, W., & Dong, J. (2022) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr56361.2022.9956333
“[18] uses SVMs and Random Forests to classify forged facial images, which is the first work to use machine-learning methods on image forensics.”
MethodLearning to Detect Deepfakes via Adaptive Attention and Constrained Difference
Su, L., Wu, B., Dai, C., Luo, H., & Chen, J. (2023) · Lecture notes in computer science · doi:10.1007/978-981-99-7356-9_31
GAN-generated face detection by color feature and ResNet
Zhang, L. (2023) · International Conference on Algorithms, High Performance Computing and Artificial Intelligence (AHPCAI) · doi:10.1117/12.3011989
“For instance, Zhang et al. [2] proposed a technique focused on detecting swapped faces.”
MethodThe recognition and analysis of fake images based on deep neural network
Luo, Q. (2024) · Applied and Computational Engineering · doi:10.54254/2755-2721/40/20230663
DeepFake Image Detection Using Adaptive Discriminator Augmentation (ADA)
Talib, D. A., & Abed, A. A. (2023) · International Conference on Advanced Engineering and Technologies (ICONNIC) · doi:10.1109/iconnic59854.2023.10467610
“Other methods have included analyzing image features in the frequency domain by training classifiers on the discrete Fourier transforms of the images [41] and using bag-of-words to highlight image data and passing those features to statistical classifiers like SVMs[42].”
MethodVideo Morphing Attack Detection Using Convolutional Neural Networks On Deep Fake Detection Algorithm
Boovaneswari, S., & RS, S. N. (2024) · doi:10.53555/kuey.v30i5.3495
Unmasking Deepfakes - Harnessing the Potential of 2D and 3D Convolutional Neural Network Ensembles
Bakliwal, A., Joshi, A. D., Deo, N., & Sawant, S. T. (2024) · International Conference on Innovative Trends in Information Technology (ICITIIT) · doi:10.1109/icitiit61487.2024.10580618 · Cited by 1
“Other studies utilized SVM-based classification, the former leveraging a pixel anomaly-based feature extraction technique and the latter employing the Speeded Up Robust Features (SURF) algorithm with the Bag of Words (BoW) model for detecting face swaps [8].”
MethodVideo Matting Tamper Detection Based on Compressed Tracking
Chen, Z., & Yang, X. (2024) · IEEE International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA) · doi:10.1109/iciba62489.2024.10868110
Deepfake Multi-Faces Detection Using Haar Cascades with Sift Matching AlgorithmPreprint
Al-Sanjary, O. I., Ahmed, A. A., & Al-Qaba, A. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5134415
A., M., Pushpitha Shree, P., & Brindha, G. (2024) · International Conference on IoT, Communication and Automation Technology (ICICAT) · doi:10.1109/icicat62666.2024.10923429
Gautam, A., & Singh, A. K. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-97-7384-8_44
Steinebach, M., & Frühwein, M. (2025) · Electronic Imaging · doi:10.2352/ei.2025.37.4.mwsf-300
“Already in 2017 Zhang et a. utilized SURF features combined with machine learning for face swap detection [12].”
Research on Self-Evolving Deepfake Detection Technology Based on LLM
Yoon, C., Cho, J., & Kang, J.-M. (2025) · IEEE International Conference on Consumer Technology (ICCT-Pacific) · doi:10.1109/icct-pacific63901.2025.11012805 · Cited by 1
Advances in Deepfake Detection: A Focus on Emerging Datasets and Deep Learning Techniques
Zheng, H., & Patel, R. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-94153-5_31
Detecting Forgery in War Footage for Information Accuracy
Afzal, S., Sharafudeen, M., & S, V. C. S. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-4008-9_38
Annapureddy, V. R. (2025) · International Conference on Networks and Soft Computing (ICNSoC) · doi:10.1109/icnsoc66817.2025.00063
Is It Certainly a Deepfake? Reliability Analysis in Detection & Generation Ecosystem
Köse, N., Rhodes, A. D., Çiftçi, U. A., & Demir, İ. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00035 · Cited by 5
“Deepfake detectors first focused on artifacts of fakery , learning directly from data with “blind” detectors [8, 10, 13, 29, 30, 36, 45, 51, 64, 78, 79, 79].”
Song, Y., Kuang, J., & Xu, R. (2026) · IEEE Access · doi:10.1109/access.2026.3669419 · Cited by 1
“For example, methods detecting unnatural edge transitions [2] and analyzing image spectral characteristics [3] achieved notable success in identifying inconsistencies introduced by early forgery techniques.”
Deepfake Generation and Detection: A Comprehensive Survey
Sharma, A., Zala, B., Jatan, S. A., Nahar, T., Dubey, V. G., & Parmar, C. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-19300-1_7
Liu, Z., Ni, Y., & Mou, Y. (2026) · Emerging Media · doi:10.1177/27523543261471830 · Cited by 1
“This technique has been widely implemented in commercial applications owing to its user-friendly nature and its capacity to generate results within seconds (Zhang et al., 2017).”
Metadata
- Authors
- Zhang, Y., Zheng, L., & Thing, V. L. L.
- Affiliations
- Institute for Infocomm Research (SG)
- Venue
- 2017 IEEE 2nd International Conference on Signal and Image Processing (ICSIP)
- Publisher
- IEEE
- Published
- 2017-08
- Pages
- 15-19
- OpenAlex
- W2774037436
- Cited by
- 154
- References
- 14
- Citations by year
- 2026: 5 · 2025: 27 · 2024: 26 · 2023: 26 · 2022: 30 · 2021: 19 · 2020: 13 · 2019: 4 · 2018: 4
- FWCI
- 1.68
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Signal and Image Processing (ICSIP) (publishedVersion)
- Funders
- National Research Foundation Singapore
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Physical Sciences › Computer Science › Signal Processing › Biometric Identification and Security
- Keywords
- face swapping, deepfake detection, machine learning detection
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.