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
.
Filters
CNN and RNN using Deepfake detection
2024
International Journal of Science and Research Archive
Deep fake Detection is the task of detecting the fake images that have been generated using deep learning techniques. ...
The goal of deep fake detection is to identify such manipulations and distinguish them from real videos or images. ...
In addition to the traditional deepfake detection models, a hybrid approach was introduced to effectively detect the fake images for example proposed a two-stream network for detecting face tampering. ...
doi:10.30574/ijsra.2024.11.2.0460
fatcat:nlw63ltmxjggxbnxmr2q5m5aji
M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection
[article]
2022
arXiv
pre-print
M2TR further learns to detect forgery artifacts in the frequency domain to complement RGB information through a carefully designed cross modality fusion block. ...
In addition, to stimulate Deepfake detection research, we introduce a high-quality Deepfake dataset, SR-DF, which consists of 4,000 DeepFake videos generated by state-of-the-art face swapping and facial ...
CONCLUSION In this paper, we presented a two-stream network Multi-modal Multi-scale Transformer (M2TR) for Deepfake detection, which uses multi-scale transformers to capture subtle local inconsistency ...
arXiv:2104.09770v3
fatcat:hskipz7oxfcdrnix6b4rdgwnai
M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection
2022
Proceedings of the 2022 International Conference on Multimedia Retrieval
In addition, to stimulate Deepfake detection research, we introduce a high-quality Deepfake dataset, SR-DF, which consists of 4,000 DeepFake videos generated by state-of-the-art face swapping and facial ...
M2TR further learns to detect forgery artifacts in the frequency domain to complement RGB information through a carefully designed cross modality fusion block. ...
CONCLUSION In this paper, we presented a two-stream network Multi-modal Multi-scale Transformer (M2TR) for Deepfake detection, which uses multi-scale transformers to capture subtle local inconsistency ...
doi:10.1145/3512527.3531415
fatcat:74t4bc54rvhchmbiutbspcy2ui
FakeOut: Leveraging Out-of-domain Self-supervision for Multi-modal Video Deepfake Detection
[article]
2024
arXiv
pre-print
Thus, there is a pressing need for accurate and robust deepfake detection methods, that can detect forgery techniques not seen during training. ...
In this work, we explore whether this can be done by leveraging a multi-modal, out-of-domain backbone trained in a self-supervised manner, adapted to the video deepfake domain. ...
In order to construct positive training pairs across two modalities, two stream are sampled from the same location of a video. ...
arXiv:2212.00773v2
fatcat:t4tw2gaisnh4rdvdfmow3wm45m
Deepfakes Detection Techniques Using Deep Learning: A Survey
2021
Journal of Computer and Communications
This work primarily focuses on providing a comprehensive study for deepfake detection using deep-learning methods such as Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), and Long short-term ...
As a result, many deep learning approaches such as long short-term memory (LSTM), recurrent neural network (RNN) and even the hybrid approaches has been proposed to in order to detect deepfakes images ...
[29] for example proposed a two-stream network for detecting face tampering (see Figure 3 ). ...
doi:10.4236/jcc.2021.95003
fatcat:4zpksmozhrdmjh6e25l2g3jjru
Fighting Malicious Media Data: A Survey on Tampering Detection and Deepfake Detection
[article]
2022
arXiv
pre-print
Accordingly, the means of defense include image tampering detection and Deepfake detection, which share a wide variety of properties. ...
In this paper, we provide a comprehensive review of the current media tampering detection approaches, and discuss the challenges and trends in this field for future research. ...
More specifically, [277] detects tampered face images with a two-stream network, where a face classification stream captures the forgery artifacts and a patch triplet stream recognizes noise residual ...
arXiv:2212.05667v1
fatcat:iuqs2d3qhbay3gryewjlbk75jm
Efficient Temporally-Aware DeepFake Detection using H.264 Motion Vectors
[article]
2024
arXiv
pre-print
This could lead to new, real-time temporally-aware DeepFake detection methods for video calls and streaming. ...
Most current DeepFake detection methods analyze each frame independently, ignoring inconsistencies and unnatural movements between frames. ...
Finally for our two-stream network, we combine the two different modalities of input, RGB and MVs by concatenating the last layers of their respective MobileNets, which consist of a single value. ...
arXiv:2311.10788v2
fatcat:2itl477a5jf6pmmq3puylrvmva
MC-LCR: Multi-modal contrastive classification by locally correlated representations for effective face forgery detection
[article]
2022
arXiv
pre-print
To address such limitations, we propose a novel framework named Multi-modal Contrastive Classification by Locally Correlated Representations(MC-LCR), for effective face forgery detection. ...
It helps the network learn more discriminative and generalized representations. ...
In our experiments, we only use two Mixer layers as a compromise between detection performance and network scale. AT and PT are input into two MLP-Mixer networks. ...
arXiv:2110.03290v2
fatcat:kcpx3ndznve67f5rvzxz6djfj4
Countering Malicious DeepFakes: Survey, Battleground, and Horizon
[article]
2022
arXiv
pre-print
detailed interactions between the adversaries (DeepFake generation) and the defenders (DeepFake detection). ...
We present the taxonomy of various DeepFake generation methods and the categorization of various DeepFake detection methods, and more importantly, we showcase the battleground between the two parties with ...
Two vectors from the aforementioned two networks are compared for detecting the identity-to-identity discrepancies. This approach also has a good generalization ability across GANs. ...
arXiv:2103.00218v3
fatcat:ufeslcp23rghhmx474u25acoje
Voice-Face Homogeneity Tells Deepfake
[article]
2022
arXiv
pre-print
Detecting forgery videos is highly desirable due to the abuse of deepfake. ...
paper, we propose to perform the deepfake detection from an unexplored voice-face matching view. ...
Later studies take into consideration the interactions among different modalities. For instance, Wen et al. ...
arXiv:2203.02195v3
fatcat:3dwi4owonvftzpbpjitvo6sheq
Combating Online Misinformation Videos: Characterization, Detection, and Future Directions
[article]
2023
arXiv
pre-print
Based on the characterization, we systematically review existing works for detection from features of various modalities to techniques for clue integration. ...
With information consumption via online video streaming becoming increasingly popular, misinformation video poses a new threat to the health of the online information ecosystem. ...
[27] propose a two-stream method by analyzing the frame-level and temporality-level of compressed deepfake videos for detection. ...
arXiv:2302.03242v2
fatcat:rbahcqlopbanzozf3marhv44fi
2020 Index IEEE Journal of Selected Topics in Signal Processing Vol. 14
2020
IEEE Journal on Selected Topics in Signal Processing
., +, JSTSP May 2020 676-687 Interactive systems A Multi-Stream Recurrent Neural Network for Social Role Detection in Multiparty Interactions. ...
., +, JSTSP Feb. 2020 261-271 Robust Detection of Image Operator Chain With Two-Stream Convolutional Neural Network. ...
doi:10.1109/jstsp.2020.3029672
fatcat:6twwzcqpwzg4ddcu2et75po77u
GOTCHA: Real-Time Video Deepfake Detection via Challenge-Response
[article]
2024
arXiv
pre-print
However, existing deepfake detection techniques are asynchronous and hence ill-suited for RTDFs. ...
The findings underscore the promising potential of challenge-response systems for explainable and scalable real-time deepfake detection in practical scenarios. ...
Two linear neural network heads (for classification and regression) were trained in a self-supervised way using binary cross-entropy and LPIPS [37] as loss functions, respectively. ...
arXiv:2210.06186v3
fatcat:epiwybbuxndt5glvvseyidvmte
Deepfake Generation and Detection: A Benchmark and Survey
[article]
2024
arXiv
pre-print
In addition to the advancements in deepfake generation, corresponding detection technologies need to continuously evolve to regulate the potential misuse of deepfakes, such as for privacy invasion and ...
We closely follow the latest developments in https://github.com/flyingby/Awesome-Deepfake-Generation-and-Detection. ...
backbone network for Deepfake detection by designing air-frequency interaction convolution. ...
arXiv:2403.17881v1
fatcat:iqies3vnbne5tg26f7y26dtscq
Joint Engagement Classification using Video Augmentation Techniques for Multi-person Human-robot Interaction
[article]
2022
arXiv
pre-print
Affect understanding capability is essential for social robots to autonomously interact with a group of users in an intuitive and reciprocal way. ...
affect recognition in the multi-person human-robot interaction in the wild. ...
Our proposed framework can also be expanded to account for individual differences in affect across dyads by adding a deep neural network layer as the final layer trained on individual human groups. ...
arXiv:2212.14128v1
fatcat:wgshdzufmjg53mykezdatmhfbi
« Previous
Showing results 1 — 15 out of 132 results