Media Forensics and DeepFakes: An Overview
Open accessThe DOI matches its Crossref record; 289 of the 289 references it cites were found in a registry or an index.
Abstract
With the rapid progress in recent years, techniques that generate and manipulate multimedia content can now provide a very advanced level of realism. The boundary between real and synthetic media has become very thin. On the one hand, this opens the door to a series of exciting applications in different fields such as creative arts, advertising, film production, and video games. On the other hand, it poses enormous security threats. Software packages freely available on the web allow any individual, without special skills, to create very realistic fake images and videos. These can be used to manipulate public opinion during elections, commit fraud, discredit or blackmail people. Therefore, there is an urgent need for automated tools capable of detecting false multimedia content and avoiding the spread of dangerous false information. This review paper aims to present an analysis of the methods for visual media integrity verification, that is, the detection of manipulated images and videos. Special emphasis will be placed on the emerging phenomenon of deepfakes, fake media created through deep learning tools, and on modern data-driven forensic methods to fight them. The analysis will help highlight the limits of current forensic tools, the most relevant issues, the upcoming challenges, and suggest future directions for research.
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.
Integrity signals
Registry
Cites
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.
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?
The cited source doesn’t support the claim as written.
In the citing paper
“The work does not define a method category and does not compare parameters between methods.”
From Deepfake detection using deep learning methods: A systematic and comprehensive review (2024) · cites this paper as L. Verdoliva (2020)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“With the majority of the surveyed papers being conventional media forensics approaches, the overlapping is insignificant.”
From Countering Malicious DeepFakes: Survey, Battleground, and Horizon (2022) · cites this paper as Verdoliva (2020)
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“according to the consensus in the DeepFake field (Mirsky and Lee 2021; Verdoliva 2020; Tolosana et al. 2020), there are mainly four categories based on their function: entire face synthesis, attribute manipulation, identity swap, and expression swap”
From Countering Malicious DeepFakes: Survey, Battleground, and Horizon (2022) · cites this paper as Verdoliva 2020
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“according to the surveys (Verdoliva 2020; Nguyen et al. 2019d; Tolosana et al. 2020; Lyu 2020), GANs are the mainstream baseline technique.”
From Countering Malicious DeepFakes: Survey, Battleground, and Horizon (2022) · cites this paper as Verdoliva 2020
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Deepfakes have already been exploited in contexts ranging from political disinformation campaigns to identity theft and personal harassment”
From Cross-dataset video deepfake detection using Transformer and CNN architectures (2026) · cites this paper as [2]
Experimental — accuracy not yet measured
7 more claims about this paper
Create a free account to see every claim other papers make about this one.
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
Items 201–289 of 289
RelGAN: Multi-Domain Image-to-Image Translation via Relative AttributesPreprint
Wu, P.-W., Lin, Y., Chang, C.-H., Chang, E. Y., & Liao, S.-W. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.07269 · Cited by 83 · cited 1×
“These include content generation [22], [23], image/video synthesis from semantic labels [24], [25] or sketches [26] or text [27], changes of style and attributes [22], [28], [29], domain translation [30], up to expression transfer [25], face swapping typical of deepfakes [31] and talking-head video editing [32].”
BackgroundDetecting image splicing in the wild (WEB)
Zampoglou, M., Papadopoulos, S., & Kompatsiaris, Y. (2015) · 2015 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew.2015.7169839 · Cited by 82 · cited 2×
“Wild Web [213] 2015 real-world cases 90 / 9,657 72×45 − 3000×2222 PNG, BMP, JPG, GIF”
Median Filtered Image Restoration and Anti-Forensics Using Adversarial Networks
Kim, D., Jang, H., Mun, S.-M., Choi, S., & Lee, H.-K. (2017) · IEEE Signal Processing Letters · doi:10.1109/lsp.2017.2782363 · Cited by 81 · cited 1×
“[246] a GAN-based architecture is proposed to hide the traces of median filtering in order to fool state-of-the-art CNN-based detectors.”
MethodContent-aware detection of JPEG grid inconsistencies for intuitive image forensics
Iakovidou, C., Zampoglou, M., Papadopoulos, S., & Kompatsiaris, I. Y. (2018) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2018.05.011 · Cited by 81 · cited 1×
“[70], [71], some of which even in recent years [72].”
BackgroundA feature-based approach for image tampering detection and localization
Verdoliva, L., Cozzolino, D., & Poggi, G. (2014) · 2014 IEEE International Workshop on Information Forensics and Security (WIFS) · doi:10.1109/wifs.2014.7084319 · Cited by 77 · cited 1×
“Therefore, in [107] it was proposed to extract local descriptors from same-model noise residuals to build a reference statistical”
BackgroundSelf-citationAccurate and Efficient Image Forgery Detection Using Lateral Chromatic Aberration
Mayer, O., & Stamm, M. C. (2018) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2018.2799421 · Cited by 74 · cited 1×
“efficient estimation of local displacements, are proposed in [50] and more recently in [51].”
BackgroundA Counter-Forensic Method for CNN-Based Camera Model Identification
Guera, D., Wang, Y., Bondi, L., Bestagini, P., Tubaro, S., & Delp, E. J. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2017.230 · Cited by 73 · cited 1×
“Indeed, attacking CNNs seems to be easier and more effective [254].”
BackgroundImage Provenance Analysis at Scale
Moreira, D., Bharati, A., Brogan, J., Pinto, A., Parowski, M., Bowyer, K. W., Flynn, P. J., Rocha, A., & Scheirer, W. J. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2865674 · Cited by 73
Photo forensics from JPEG dimples
Agarwal, S., & Farid, H. (2017) · 2017 IEEE Workshop on Information Forensics and Security (WIFS) · doi:10.1109/wifs.2017.8267641 · Cited by 70 · cited 1×
“In [81] model-specific JPEG features have been defined, the JPEG”
MethodImaging Sensor Noise as Digital X-Ray for Revealing Forgeries
Chen, M., Fridrich, J., Lukáš, J., & Goljan, M. (2007) · Lecture notes in computer science · doi:10.1007/978-3-540-77370-2_23 · Cited by 68 · cited 1×
“[101] a predictor is designed to adapt the decision threshold to the local image statistics, while in [102] disturbing non-”
BackgroundLearned Forensic Source Similarity for Unknown Camera Models
Mayer, O., & Stamm, M. C. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8462585 · Cited by 68
Detection of Fake Images Via The Ensemble of Deep Representations from Multi Color Spaces
He, P., Li, H., & Wang, H. (2019) · 2019 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip.2019.8803740 · Cited by 67 · cited 1×
“Therefore, to boost performance, the scores obtained at frame level can be aggregated [177], or else an ensembling approach can be applied, using different color spaces as input [186].”
MethodMake a Face: Towards Arbitrary High Fidelity Face Manipulation
Qian, S., Lin, K.-Y., Wu, W., Liu, Y., Wang, Q., Shen, F., Qian, C., & He, R. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.01013 · Cited by 67 · cited 1×
“Methods have been proposed to change the expression of a face [48], [49],”
BackgroundSatellite Image Forgery Detection and Localization Using GAN and One-Class Classifier
Yarlagadda, S. K., Güera, D., Bestagini, P., Maggie Zhu, F., Tubaro, S., & Delp, E. J. (2018) · Electronic Imaging · doi:10.2352/issn.2470-1173.2018.07.mwsf-214 · Cited by 66
Exposing Fake Images With Forensic Similarity Graphs
Mayer, O., & Stamm, M. C. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.3001516 · Cited by 65 · cited 1×
“This work has been recently extended in [161] introducing a graph-based representation that better captures the forensic relationships”
BackgroundDeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery DetectionPreprint
Jiang, L., Li, R., Wu, W., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.03024 · Cited by 65
Generative Adversarial Attacks Against Deep- Learning-Based Camera Model Identification
Chen, C., Zhao, X., & Stamm, M. C. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2945198 · Cited by 64
On the vulnerability of deep learning to adversarial attacks for camera model identification
Marra, F., Gragnaniello, D., & Verdoliva, L. (2018) · Signal Processing Image Communication · doi:10.1016/j.image.2018.04.007 · Cited by 62 · cited 1×
“but also adversarial noise, reducing the effectiveness of such attacks [245].”
BackgroundSelf-citationAdversarial Multimedia Forensics: Overview and Challenges Ahead
Barni, M., Stamm, M. C., & Tondi, B. (2018) · 2018 26th European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8553305 · Cited by 62
Physiologically-based detection of computer generated faces in video
Conotter, V., Bodnari, E., Boato, G., & Farid, H. (2014) · 2014 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip.2014.7025049 · Cited by 61 · cited 2×
“In [163] and in [164] deepfakes are revealed by the lack of variations induced by heart beating, an idea already exploited in [160] for computer generated faces.”
BackgroundMethodDeep Learning for Detecting Processing History of Images
Boroumand, M., & Fridrich, J. (2018) · Electronic Imaging · doi:10.2352/issn.2470-1173.2018.07.mwsf-213 · Cited by 60 · cited 3×
“Some methods [145], [146] train the CNN to extract compact features at”
BackgroundMethodA PUF-Based Data-Device Hash for Tampered Image Detection and Source Camera Identification
Zheng, Y., Cao, Y., & Chang, C.-H. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2926777 · Cited by 58 · cited 1×
“However, it is important to underline that active methods may be very effective in ensuring the digital integrity of media assets, and there is a growing interest on these authentication methods to protect media content [19], [20].”
BackgroundChung, J. S., Jamaludin, A., & Zisserman, A. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.02966 · Cited by 58 · cited 1×
“animated based on an audio input sequence [46], [47].”
BackgroundJiang, X., He, P., Sun, T., Xie, F., & Wang, S. (2017) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2017.2745687 · Cited by 56 · cited 1×
“Some features have been devised to detect specific artifacts, especially those generated by double JPEG compression [108], [102], [109] or related to the camera response function (CRF) [110], [111], [112].”
BackgroundIdentification of cut & paste tampering by means of double-JPEG detection and image segmentation
Barni, M., Costanzo, A., & Sabatini, L. (2010) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas.2010.5537505 · Cited by 54 · cited 1×
“[82] and localization [83], [84] methods have been proposed for both cases.”
MethodGuided filtering for PRNU-based localization of small-size image forgeries
Chierchia, G., Cozzolino, D., Poggi, G., Sansone, C., & Verdoliva, L. (2014) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2014.6854802 · Cited by 54 · cited 1×
“Further variations rely on the use of guided filtering [112], discriminative random fields [113] and multiscale analysis [15].”
MethodSelf-citationA video forensic framework for the unsupervised analysis of MP4-like file container
Iuliani, M., Shullani, D., Fontani, M., Meucci, S., & Piva, A. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2018.2859760 · Cited by 54 · cited 1×
“useless, otherwise they would provide precious information towards authenticity verification both for images and videos [15], [16], [17].”
BackgroundWhere and Who? Automatic Semantic-Aware Person Composition
Tan, F., Bernier, C., Cohen, B., Ordóñez, V., & Barnes, C. (2018) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2018.00170 · Cited by 54
SpoC: Spoofing Camera Fingerprints
Cozzolino, D., Thies, J., Rossler, A., Niesner, M., & Verdoliva, L. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00110 · Cited by 54 · cited 1×
“Instead [249] proposes a GAN-based architecture with a twofold goal, to inject traces of a real camera in synthetic images and, at the same time, reduce peculiar traces of GAN generation.”
BackgroundSelf-citationIdentification of Motion-Compensated Frame Rate Up-Conversion Based on Residual Signals
Ding, X., Yang, G., Li, R., Zhang, L., Li, Y., & Sun, X. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2676162 · Cited by 51 · cited 1×
“A similar idea is extended to videos in [72]–[74] where the noise residuals of consecutive frames are analyzed and suitable features are extracted to discover traces of both intra-frame and inter-frame manipulations.”
BackgroundThe Video Authentication and Camera Identification Database: A New Database for Video Forensics
Hosler, B. C., Zhao, X., Mayer, O., Chen, C., Shackleford, J., & Stamm, M. C. (2019) · IEEE Access · doi:10.1109/access.2019.2922145 · Cited by 51 · cited 1×
“Finally, this year the video-ACID database has been published [243], with over 12 000 videos from 46 physical devices of 36 different models.”
MethodInternet image archaeology: automatically tracing the manipulation history of photographs on the web
Kennedy, L., & Chang, S.-F. (2008) · Proceedings of the 16th ACM international conference on Multimedia · doi:10.1145/1459359.1459406 · Cited by 50 · cited 1×
“versions of the same asset, it is possible to build its manipulation history (image and video phylogeny) [9]–[11].”
BackgroundStatistical Detection of JPEG Traces in Digital Images in Uncompressed Formats
Pasquini, C., Boato, G., & Pérez‐González, F. (2017) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2017.2725201 · Cited by 47 · cited 1×
“Other methods [78], [79] look for anomalies in the statistical distribution of original DCT samples, assumed”
MethodAdversarial Scene Editing: Automatic Object Removal from Weak Supervision
Shetty, R., Fritz, M., & Schiele, B. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.01911 · Cited by 46 · cited 1×
“Besides conventional manipulations, like composition [18], copy-move [19], object removal [20], inpainting [21], a large number of new tasks can be performed.”
BackgroundCamera-based Image Forgery Localization using Convolutional Neural Networks
Cozzolino, D., & Verdoliva, L. (2018) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8553581 · Cited by 45
Self-citationVazquez-Padin, D., Fontani, M., Shullani, D., Perez-Gonzalez, F., Piva, A., & Barni, M. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2951313 · Cited by 45 · cited 1×
“improved to detect traces of inter-frame tampering, and more recently in [86] to deal with video sequences that contain bidirectional frames.”
BackgroundVideo Phylogeny: Recovering near-duplicate video relationships
Dias, Z., Rocha, A. D. R., & Goldenstein, S. K. (2011) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs.2011.6123127 · Cited by 44 · cited 1×
“Moreover, by identifying the connections among the various versions of the same asset, it is possible to build its manipulation history (image and video phylogeny) [9], [10].”
BackgroundCNN-generated images are surprisingly easy to spot... for nowPreprint
Wang, S.-Y., Wang, O., Zhang, R., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.11035 · Cited by 44 · cited 1×
“Instead, in [197] a careful pre- and postprocessing and data augmentation are applied to improve transferability.”
MethodA New Dataset for Source Identification of High Dynamic Range Images
Shaya, O. A., Yang, P., Ni, R., Zhao, Y., & Piva, A. (2018) · Sensors · doi:10.3390/s18113801 · Cited by 43 · cited 1×
“SDR (Standard Dynamic Range) and HDR (High Dynamic Range) images has been presented in [240].”
BackgroundLong, C., Basharat, A., & Hoogs, A. J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.10762 · Cited by 43
On the Detection of Digital Face ManipulationPreprint
Dang, H., Liu, F., Stehouwer, J., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1910.01717 · Cited by 41
Poster: Towards Robust Open-World Detection of Deepfakes
Sohrawardi, S. J., Chintha, A., Thai, B., Seng, S., Hickerson, A., Ptucha, R., & Wright, M. (2019) · Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3319535.3363269 · Cited by 41 · cited 1×
“ve, they perform face manipulation on a frame-by-frame basis and hence may incorrectly follow the face movements. Several methods have been proposed in the literature to exploit this point. In [187], [188] a convolutional Long Short Term Memory (LSTM) network is used to exploit such dependencies and improve upon singl…”
MethodGloe, T., Borowka, K., & Winkler, A. (2010) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.839034 · Cited by 40 · cited 1×
“efficient estimation of local displacements, are proposed in [50] and more recently in [51].”
BackgroundOn the Transferability of Adversarial Examples against CNN-based Image Forensics
Barni, M., Kallas, K., Nowroozi, E., & Tondi, B. (2019) · ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2019.8683772 · Cited by 40
Detecting and Simulating Artifacts in GAN Fake ImagesPreprint
Zhang, X., Karaman, S., & Chang, S.-F. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.06515 · Cited by 39 · cited 1×
“In [174], instead, a GAN simulator is proposed to reproduce common GAN-image artifacts, which manifest as spectral peaks in the Fourier domain.”
BackgroundAnalysis of Adversarial Attacks against CNN-based Image Forgery Detectors
Gragnaniello, D., Marra, F., Poggi, G., & Verdoliva, L. (2018) · 2018 26th European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8553560 · Cited by 38
Self-citationFind it! Fraud Detection Contest Report
Artaud, C., Sidère, N., Doucet, A., Ogier, J.-M., & d’Andecy, V. P. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8545428 · Cited by 36 · cited 1×
“of all the winning approaches in several forensic challenges, like in the IEEE Forensics challenge [120], [96], the fraud detection contest on receipts [255] or the Camera Model Identification Challenge organized for the 2018 IEEE Signal Processing Cup and hosted on Kaggle [225].”
BackgroundForgery Authentication in Extreme Wide-Angle Lens Using Distortion Cue and Fake Saliency Map
Fu, H., & Cao, X. (2012) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2012.2195492 · Cited by 33 · cited 1×
“Finally, in [52] it is exploited the radial distortion that characterizes the wide-angle lens typically used for indoor/outdoor video surveillance.”
BackgroundTowards Generalizable Forgery Detection with Locality-aware AutoEncoder.Preprint
Du, M., Pentyala, S., Li, Y., & Hu, X. (2019) · arXiv (Cornell University) · Cited by 33
Discrimination Between Computer Generated And Natural Human Faces Based On Asymmetry Information
Boato, G., Dang‐Nguyen, D., & Natale, F. G. B. D. (2012) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.43224 · Cited by 32
A gradient-based pixel-domain attack against SVM detection of global image manipulations
Chen, Z., Tondi, B., Li, X., Ni, R., Zhao, Y., & Barni, M. (2017) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs.2017.8267668 · Cited by 29 · cited 1×
“In multimedia forensics, some early papers [238], [239] focused on attacking machine learning detectors based on the rich model handcrafted features, obtaining mixed results with high complexity.”
BackgroundWe Need No Pixels: Video Manipulation Detection Using Stream Descriptors
Güera, D., Baireddy, S., Bestagini, P., Tubaro, S., & Delp, E. J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.08743 · Cited by 29 · cited 1×
“integrity, and that a skilled attacker modified metadata to make them useless, otherwise they would provide precious information towards authenticity verification both for images and videos [15], [16], [17]. On the other hand, it is worth noting that metadata are routinely canceled when media assets are uploaded Splicin…”
BackgroundDetecting Photoshopped Faces by Scripting PhotoshopPreprint
Wang, S.-Y., Wang, O., Owens, A., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.05856 · Cited by 27 · cited 1×
“Finally, in [134] a CNN based solution is devised to detect artifacts introduced by a specific Photoshop tool, Face-Aware Liquify, which performs image warping of human faces.”
MethodHow Relevant Is the Turing Test in the Age of Sophisbots?
Boneh, D., Grotto, A. J., McDaniel, P., & Papernot, N. (2019) · IEEE Security & Privacy · doi:10.1109/msec.2019.2934193 · Cited by 27 · cited 1×
“e authentication methods [259], [260]. In past decades, a large body of research was produced on digital watermarking [261]. There is now much interest in blockchain technology [262], in criptography [263], and even new active methods have been proposed to ensure the integrity of digital media [264] or to protect indi…”
BackgroundDurall, R., Keuper, M., & Keuper, J. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00791 · Cited by 26 · cited 1×
“fail to reproduce spectral distributions of natural training data correctly [174].”
BackgroundSource Generator Attribution via Inversion
Albright, M., & McCloskey, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.02259 · Cited by 24
A Possible Pitfall in the Experimental Analysis of Tampering Detection Algorithms
Cattaneo, G., & Roscigno, G. (2014) · International Conference on Network-Based Information Systems · doi:10.1109/nbis.2014.82 · Cited by 23 · cited 1×
“Nonetheless, it exhibits a strong polarization, highlighted in [203].”
BackgroundTampered Speaker Inconsistency Detection with Phonetically Aware Audio-visual Features
Korshunov, P., Halstead, M., Castán, D., Graciarena, M., McLaren, M., Burns, B., Lawson, A., & Marcel, S. (2019) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · Cited by 23
Korus, P., & Memon, N. D. (2019) · 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/cvpr.2019.00882 · Cited by 21 · cited 1×
“criptography [263], and even new active methods have been proposed to ensure the integrity of digital media [264] or to protect individuals from becoming the victims of AI attacks [265].”
BackgroundFace X-ray for More General Face Forgery DetectionPreprint
Li, L., Bao, J., Zhang, T., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.13458 · Cited by 21 · cited 1×
“pletely different perspective is followed in [199].”
Multiple Classifier Systems for Image Forgery Detection
Cozzolino, D., Gargiulo, F., Sansone, C., & Verdoliva, L. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-41184-7_27 · Cited by 20 · cited 1×
“In [269] several forensic tools, based on complementary hypotheses, are fused following various strategies, with results confirming that”
BackgroundSelf-citationTwo-Stream Network for Detecting Double Compression of H.264 Videos
Nam, S.-H., Park, J., Kim, D., Yu, I.-J., Kim, T., & Lee, H.-K. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8802966 · Cited by 20 · cited 1×
“264 video analysis in [125], where a two-stream neural network is proposed to analyze separately intra-coded frames and predictive frames.”
Image splicing detection based on noncausal Markov model
Zhao, X., Wang, S., Li, S., Li, J., & Yuan, Q. (2013) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2013.6738919 · Cited by 19 · cited 1×
“Therefore, the most effective features are typically extracted from noise residuals, either in the spatial [114], [115] or in a transform domain [116], [117].”
BackgroundResidual-based forensic comparison of video sequences
Mullan, P., Cozzolino, D., Verdoliva, L., & Rieß, C. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8296533 · Cited by 19 · cited 1×
“A similar idea is extended to videos in [64], [65], [66] where the noise residuals of consecutive frames are analyzed and suitable features are extracted to discover traces of both intra-frame and inter-frame manipulations.”
BackgroundSelf-citationZooming into Face Forensics: A Pixel-level AnalysisPreprint
Li, J., Shen, T., Zhang, W., Ren, H., Zeng, D., & Mei, T. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.05790 · Cited by 19 · cited 1×
“using the approach described in [182], however this analysis can be accomplished in different ways [183].”
MethodThe PS-Battles Dataset - an Image Collection for Image Manipulation DetectionPreprint
Heller, S., Rossetto, L., & Schuldt, H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1804.04866 · Cited by 19
Camera response function signature for digital forensics - Part II: Signature extraction
Ng, T.-T. (2009) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs.2009.5386461 · Cited by 18 · cited 1×
“Some features have been devised to detect specific artifacts, especially those generated by double JPEG compression [108], [102], [109] or related to the camera response function (CRF) [110], [111], [112].”
BackgroundHiding Faces in Plain Sight: Disrupting AI Face Synthesis with Adversarial Perturbations
Li, Y., Yang, X., Wu, B., & Lyu, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.09288 · Cited by 18 · cited 1×
“criptography [263], and even new active methods have been proposed to ensure the integrity of digital media [264] or to protect individuals from becoming the victims of AI attacks [265].”
BackgroundCounter-forensics in machine learning based forgery detection
Marra, F., Poggi, G., Roli, F., Sansone, C., & Verdoliva, L. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2182173 · Cited by 18 · cited 1×
“In multimedia forensics, some early papers [252], [253] focused on attacking machine learning detectors based on the rich”
BackgroundSelf-citationUnsupervised fusion for forgery localization exploiting background information
Ferrara, P., Fontani, M., Bianchi, T., Rosa, A. D., Piva, A., & Barni, M. (2015) · IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew.2015.7169770 · Cited by 17 · cited 2×
“o a detector score, one should take into account its overall reliability, the level of confidence associated with the specific decision and its compatibility with other detectors. Experiments in [250], [251] show that the DST approach, with reasonable fusion rules, outperforms consistently all individual tools and also…”
BackgroundMethodThe PS-Battles Dataset - an Image Collection for Image Manipulation Detection
Heller, S., Rossetto, L., & Schuldt, H. (2018) · edoc (University of Basel) · doi:10.5281/zenodo.7467932 · Cited by 17
PRNU-Based Forgery Localization in a Blind Scenario
Cozzolino, D., Marra, F., Poggi, G., Sansone, C., & Verdoliva, L. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-68548-9_52 · Cited by 16 · cited 1×
“extended to blind scenarios, where no prior information about the camera is known provided a suitable clustering procedure identifies the images which share the same PRNU [104], [114].”
BackgroundSelf-citation3D-Model-Based Video Analysis for Computer Generated Faces Identification
Dang‐Nguyen, D., Boato, G., & Natale, F. G. B. D. (2015) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2015.2427778 · Cited by 15 · cited 1×
“For example, in [187] detection relies on the spatial-temporal deformations of a 3D model that fits the face.”
BackgroundExposing GAN-synthesized Faces Using Landmark LocationsPreprint
Yang, X., Li, Y., Qi, H., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.00167 · Cited by 14 · cited 2×
“Other detection methods rely on face warping artifacts [165], face landmark locations [166] or head pose inconsistencies [167].”
BackgroundExploiting Human Social Cognition for the Detection of Fake and Fraudulent Faces via Memory NetworksPreprint
Fernando, T., Fookes, C., Denman, S., & Sridharan, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.07844 · Cited by 13 · cited 1×
“In [198] this is achieved using hierarchical neural memory networks.”
MethodAttacking the triangle test in sensor-based camera identification
Marra, F., Roli, F., Cozzolino, D., Sansone, C., & Verdoliva, L. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7026074 · Cited by 12 · cited 1×
“detectors [247], and then more powerful attacks [248], in an arms race typical of such two-player games.”
BackgroundSelf-citationVideos versus still images: Asymmetric sensor pattern noise comparison on mobile phones
Galdi, C., Hartung, F. H., & Dugelay, J. (2017) · Electronic Imaging · doi:10.2352/issn.2470-1173.2017.7.mwsf-331 · Cited by 12 · cited 1×
“Another mixed dataset, SOCRATES, is proposed in [228].”
BackgroundGANprintR: Improved Fakes and Evaluation of the State of the Art in Face Manipulation Detection
Neves, J. C., Tolosana, R., Vera-Rodriguez, R., Lopes, V., Proenca, H., & Fierrez, J. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.3007250 · Cited by 12
Towards Generalizable Deepfake Detection with Locality-aware AutoEncoderPreprint
Du, M., Pentyala, S., Li, Y., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.05999 · Cited by 12 · cited 1×
“An autoencoder is also used in [194], where a pixelwise mask is used to enforce the model to learn intrinsic”
MethodWu, Y., AbdAlmageed, W., & Natarajan, P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.09765 · Cited by 11 · cited 1×
“For example, when objects are copy-pasted from images available on the web, several near-identical copies can be detected [7], [8], suggesting a possible manipulation.”
BackgroundGenerate, Segment and Refine: Towards Generic Manipulation SegmentationPreprint
Zhou, P., Chen, B.-C., Han, X., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.09729 · Cited by 11 · cited 1×
“In [152] localization is pursued by first introducing a process to generate forged (harder) examples during training, so as to generalize across a large variety of possible manipulations.”
BackgroundImage Forensics from Chroma Subsampling of High-Quality JPEG Images
Lorch, B., & Rieß, C. (2019) · ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec) · doi:10.1145/3335203.3335722 · Cited by 11 · cited 1×
“Also, chroma subsampling presents specific clues due to integer rounding [82].”
BackgroundOn the generalization of GAN image forensicsPreprint
Xuan, X., Peng, B., Wang, W., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.11153 · Cited by 10
Robust Homomorphic Image Hashing.
Singh, P., & Farid, H. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · Cited by 7
Real or Fake? Spoofing State-Of-The-Art Face Synthesis Detection Systems.Preprint
Neves, J. C., Tolosana, R., Vera-Rodríguez, R., Lopes, V., & Proença, H. P. M. C. (2019) · arXiv (Cornell University) · Cited by 6
Analysis of adversarial attacks against CNN-based image forgery detectorsPreprint
Gragnaniello, D., Marra, F., Poggi, G., & Verdoliva, L. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.08426 · Cited by 4 · cited 1×
“chitectures trained for the same task [242], [243].”
BackgroundSelf-citationDetecting GAN generated errorsPreprint
Zhu, X., Che, F., Yang, T., Yu, T., Meger, D., & Dudek, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.00527 · Cited by 3
Image and Video Processing History Recovery
Bianchi, T., & Piva, A. (2015) · Wiley eBooks · doi:10.1002/9781118705773.ch15 · Cited by 1
(2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings
Research works printed but not linked: 7
- chung, 0, Proc Brit Mach Vision Conf, 1
- [36] T.-C. Wang, M.-Y. Liu, J.-Y. Zhu, G. Liu, A. Tao, J. Kautz, and B. Catanzaro, 'Video-to-video synthesis,' in Conference on Neural Information Processing Systems , 2018. -
- [59] J. Ho, O. Au, J. Zhou, and Y. Guo, 'Inter-channel demosaicing traces for digital image forensics,' in IEEE International Conference on Multimedia and Expo , 2010. -
- [133] M. Barni, Q.-T. Phan, and B. Tondi, 'Copy Move SourceTarget Disambiguation through Multi-Branch CNNs,' arXiv preprint arXiv:1912.12640v1 , 2019. - · arXiv
- [241] I. Goodfellow, J. Shlens, and C. Szegedy, 'Intriguing properties of neural networks,' in Explaining and harnessing adversarial examples , 2015. -
- [17] D. G¨ uera, S. Baireddy, P. Bestagini, S. Tubaro, and E. Delp, 'We need no pixels: Video manipulation detection using stream descriptors,' in ICML Workshops , 2019. -
- [105] S. Chakraborty and M. Kirchner, 'PRNU-based forgery detection with discriminative random fields,' in International Symposium on Electronic Imaging: Media Watermarking, Security, and Forensics , 2017. -
Cited by
Deepfakes and beyond: A Survey of face manipulation and fake detection
Tolosana, R., Vera-Rodriguez, R., Fierrez, J., Morales, A., & Ortega-Garcia, J. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.06.014 · Cited by 1,237
Deepfake Detection: A Systematic Literature Review
Rana, M. S., Nobi, M. N., Murali, B., & Sung, A. H. (2022) · IEEE Access · doi:10.1109/access.2022.3154404 · Cited by 573
Masood, M., Nawaz, M., Malik, K. M., Javed, A., Irtaza, A., & Malik, H. (2023) · Applied Intelligence · doi:10.1007/s10489-022-03766-z · Cited by 543
“In [12], the main focus was on generic image manipulation and multimedia forensic techniques.”
BackgroundInfluentialDynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstruction
Gafni, G., Thies, J., Zollhofer, M., & Niesner, M. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00854 · Cited by 501
Two-Branch Recurrent Network for Isolating Deepfakes in Videos
Masi, I., Killekar, A., Mascarenhas, R. M., Gurudatt, S. P., & AbdAlmageed, W. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58571-6_39 · Cited by 471
Metadata
- Authors
- Verdoliva, L.
- Affiliations
- University of Naples Federico II (IT)
- Corresponding author
- Verdoliva, Luisa
- ORCID
- Verdoliva, Luisa
- Venue
- IEEE Journal of Selected Topics in Signal Processing
- Published
- 2020-08
- Volume
- 14
- Issue
- 5
- Pages
- 910-932
- ISSN
- 1932-4553 (ISSN-L)
- 1941-0484
- OpenAlex
- W3002581072
- Semantic Scholar
- f4094ab2fd789d09571926868cb2420896126a0f
- Cited by
- 979
- References
- 278
- Citations by year
- 2026: 263 · 2025: 190 · 2024: 165 · 2023: 121 · 2022: 121 · 2021: 86 · 2020: 33
- FWCI
- 35.09
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Journal of Selected Topics in Signal Processing (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- IEEE RESOURCE CENTERS
- Funders
- Defense Advanced Research Projects Agency (FA8750-16-2-0204)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Steganography and Watermarking Techniques
- Keywords
- multimedia forensics, deepfakes, fake image detection, fake video detection, deep learning, misinformation, synthetic media
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.