A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Filters
Robust Deepfake On Unrestricted Media: Generation And Detection
[article]
2022
arXiv
pre-print
Preserved Fulltext
It also discusses possible ways to improve the robustness of deepfake detection for a wide variety of media (e.g., in-the-wild images and videos). ...
This chapter explores the evolution of and challenges in deepfake generation and detection. ...
Acknowledgments This research was partly supported by JSPS KAKENHI Grants (JP16H06302, JP18H04120, JP21H04907, JP20K23355, JP21K18023) and JST CREST Grants (JPMJCR20D3, JP-MJCR18A6), Japan. ...
arXiv:2202.06228v1
fatcat:a37q2lf7w5bcbekk5esmbx2goe
DeePhy: On Deepfake Phylogeny
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Model attribution helps in enhancing the explainability of the detection results by providing information on the generative model employed. ...
The results highlight the need to evolve the research of model attribution of deepfakes and generalize the process over a variety of deepfake generation techniques. ...
This research is supported through a grant from Ministry of Home Affairs, Government of India. K. Thakral is partially supported by the PMRF Fellowship. S. ...
arXiv:2209.09111v1
fatcat:vhpwrkcajbg4pbabhzhnnda7um
CapST: An Enhanced and Lightweight Model Attribution Approach for Synthetic Videos
[article]
2024
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2023 pre-print arXiv:2311.03782v1 fatcat:rvwmpoklwzhupa5mca2mxjrcmaOther Versions
The Capsule module captures intricate hierarchies among features for robust identification of deepfake attributes. ...
This study formulates Deepfakes model attribution as a multiclass classification task, proposing a segment of VGG19 as a feature extraction backbone, known for its effectiveness in imagerelated tasks, ...
FBDB (Facebook Deepfake Detection Dataset) [31] is a large-scale dataset developed by Facebook for deepfake detection research. ...
arXiv:2311.03782v3
fatcat:2u6pbojzmjdvzjmr3apnkj2rxa
Detecting Deepfake Videos using Attribution-Based Confidence Metric
2020
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose the application of the state-of-theart attribution based confidence (ABC) metric for detecting deepfake videos. ...
The deep learning model is trained only on original videos. The ABC metric uses the trained model to generate confidence values. For, original videos, the confidence values are greater than 0.94. ...
Steven Fernandes acknowledges support from the University of Central Florida Preeminent Post-doctoral Fellowship Program. ...
doi:10.1109/cvprw50498.2020.00162
dblp:conf/cvpr/FernandesREP0OV20
fatcat:fghj6srcdjhmtdfd6ngl34iiwi
DeepFake Detection for Human Face Images and Videos: A Survey
2022
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Additionally, the issue of how DeepFake detection aims to generate a generalized DeepFake detection model will be analyzed. ...
We will review the existing types of DeepFake creation techniques and sort them into five major categories. Generally, DeepFake models are trained on DeepFake datasets and tested with experiments. ...
A fine-tuning approach is employed to improve the detection capacity of existing DNN models, explore artifact clues and train DNN models on different types of datasets to improve the generalization capacity ...
doi:10.1109/access.2022.3151186
fatcat:imz6hdtofrbxfcfi6kput2mffi
FakeTagger: Robust Safeguards against DeepFake Dissemination via Provenance Tracking
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In recent years, DeepFake is becoming a common threat to our society, due to the remarkable progress of generative adversarial networks (GAN) in image synthesis. ...
While passive detection simply tells whether the image is fake or real, DeepFake provenance, on the other hand, provides clues for tracking the sources in DeepFake forensics. ...
the GAN-based image synthesis [44, 63] , instead of merely improving the generation capabilities in unknown GANs and robustness against various degradations in detecting DeepFakes passively. ...
arXiv:2009.09869v3
fatcat:eemjt2fnxrgr7eg4dursbec5hu
Towards Solving the DeepFake Problem : An Analysis on Improving DeepFake Detection using Dynamic Face Augmentation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Moreover, we also propose a general-purpose data pre-processing guideline to train and evaluate existing architectures allowing us to improve the generalizability of these models for deepfake detection ...
The creation of altered and manipulated faces has become more common due to the improvement of DeepFake generation methods. ...
A deepfake generator first learns the facial features and attributes of the subjects and then generates a forged face by selectively altering these attributes. ...
arXiv:2102.09603v3
fatcat:snykyvaprrdpzizo66z3yc5m44
TOPFORMER: Topology-Aware Authorship Attribution of Deepfake Texts with Diverse Writing Styles
[article]
2024
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We propose TopFormer to improve existing AA solutions by capturing more linguistic patterns in deepfake texts by including a Topological Data Analysis (TDA) layer in the Transformer-based model. ...
In particular, in this work, we investigate the more general version of the problem, known as Authorship Attribution (AA), in a multi-class setting–i.e., not only determining if a given text is a deepfake ...
Lastly, for the human-based approaches, to improve human detection of deepfake texts, researchers have utilized two main techniques -training [10, 42] and not training [14] . ...
arXiv:2309.12934v2
fatcat:c6mfrxx5jncffihkbfng5shaxe
Countering Malicious DeepFakes: Survey, Battleground, and Horizon
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2103.00218v1 fatcat:dgucz5mitbbh7fxoypryg6eoiiOther Versions
To fill this gap, in this paper, we provide a comprehensive overview and detailed analysis of the research work on the topic of DeepFake generation, DeepFake detection as well as evasion of DeepFake detection ...
, e.g., the evasion of DeepFake detection. ...
Second, for each DeepFake detection method, we collected the generation method detected by them and sort the generation methods by the detection accuracy/AUC (e.g., if the detection accuracy on generation ...
arXiv:2103.00218v3
fatcat:ufeslcp23rghhmx474u25acoje
Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Data
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
and model-level perturbations, (4) stays hard to be detected by adversaries, and (5) converts deepfake detection and attribution into trivial tasks and outperforms the recent state-of-the-art baselines ...
Thus, we seek a proactive and sustainable solution on deepfake detection, that is agnostic to the evolution of generative models, by introducing artificial fingerprints into the models. ...
Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the DARPA. ...
arXiv:2007.08457v7
fatcat:i3q36p2xg5hcbfuuyk7u6kpfb4
Towards mitigating uncann(eye)ness in face swaps via gaze-centric loss terms
[article]
2024
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
We additionally propose a novel loss equation for the training of face swapping models, leveraging a pretrained gaze estimation network to directly improve representation of the eyes. ...
Our method additionally reduces the prevalence of the eyes as a deciding factor when viewers perform deepfake detection tasks. ...
Ethan Wilson acknowledges funding from the University of Florida Graduate School Preeminence Award (GSPA) and Generation NEXT scholarship (Award 1833908). ...
arXiv:2402.03188v1
fatcat:zayp4nuhcvgyfnh7mlsqrajzha
Detecting Deepfake Voice Using Explainable Deep Learning Techniques
2022
Applied Sciences
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Fake media, generated by methods such as deepfakes, have become indistinguishable from real media, but their detection has not improved at the same pace. ...
In this paper, we present a human perception level of interpretability for deepfake audio detection. ...
Conclusions In this paper, we presented a general interpretation for deepfake audio detection by analyzing the attribution score patterns of several post-hoc XAI methods. ...
doi:10.3390/app12083926
fatcat:vjuwm2j4jbhtrgxgb3v6sufwya
Deepfake Videos in the Wild: Analysis and Detection
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2103.04263v1 fatcat:2rsmq6xgunbedfebocju6a7oxu
Even if detection schemes are shown to perform well on existing datasets, it is unclear how well the methods generalize to real-world deepfakes. ...
Fourth, we explore the potential for transfer learning schemes and competition-winning techniques to improve defenses. ...
Towards Improving Detection Performance In light of the poor generalization capabilities of existing detection schemes, we consider multiple approaches for improving performance, and evaluate their efficacy ...
arXiv:2103.04263v2
fatcat:sykmrrck6fciheh5xcwacam2lm
Masked Conditional Diffusion Model for Enhancing Deepfake Detection
[article]
2024
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
Extensive experiments demonstrate that forgery images generated with our method are of high quality and helpful to improve the performance of deepfake detection models. ...
It generates a variety of forged faces from a masked pristine one, encouraging the deepfake detection model to learn generic and robust representations without overfitting to special artifacts. ...
[19] propose a deepfake dataset (FaceForensics++) for training deepfake detection models. ...
arXiv:2402.00541v1
fatcat:pjvzesm5cjcx7obhk62r32mzye
Detecting and Recovering Sequential DeepFake Manipulation
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Since photorealistic faces can be readily generated by facial manipulation technologies nowadays, potential malicious abuse of these technologies has drawn great concerns. ...
This new threat requires us to detect a sequence of facial manipulations, which is vital for both detecting deepfake media and recovering original faces afterwards. ...
This expands the scope of existing deepfake problem by adding sequential manipulation information and poses a new challenge for current one-step deepfake detection methods. ...
arXiv:2207.02204v1
fatcat:6y6clp4qanbu7bat2tggzhr2ga
« Previous
Showing results 1 — 15 out of 1,197 results