MEG: Multi-Evidence GNN for Multimodal Semantic Forensics

Sabir, Ekraam; Jaiswal, Ayush; AbdAlmageed, Wael; Natarajan, Prem

doi:10.1109/icpr48806.2021.9413053

Cited by 10 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the face images in the fake videos are generated frame by frame, the Auto-Encoders does not consider the previous images when generating, which often leads to multiple anomalies in some areas between the previous and subsequent frames. Inspired by [17], Sabir et al [18] propose to combine temporal network and face preprocessing techniques to detect tampered videos. The video will destroy the continuity between adjacent frames, resulting in the above detection method not performing well under compression.…”

Section: Video-level Detection Methodsmentioning

confidence: 99%

Deepfake Video Detection Based on MesoNet with Preprocessing Module

Xia

Qiao

et al. 2022

Symmetry

View full text Add to dashboard Cite

With the development of computer hardware and deep learning, face manipulation videos represented by Deepfake have been widely spread on social media. From the perspective of symmetry, many forensics methods have been raised, while most detection performance might drop under compression attacks. To solve this robustness issue, this paper proposes a Deepfake video detection method based on MesoNet with preprocessing module. First, the preprocessing module is established to preprocess the cropped face images, which increases the discrimination among multi-color channels. Next, the preprocessed images are fed into the classic MesoNet. The detection performance of proposed method is verified on two datasets; the AUC on FaceForensics++ can reach 0.974, and it can reach 0.943 on Celeb-DF which is better than the current methods. More importantly, even in the case of heavy compression, the detection rate can still be more than 88%.

show abstract

Section: Video-level Detection Methodsmentioning

confidence: 99%

Deepfake Video Detection Based on MesoNet with Preprocessing Module

Xia

Qiao

et al. 2022

Symmetry

View full text Add to dashboard Cite

show abstract

“…Later approaches employ deep learning (DL) to capture high-level forgery features. Most DL-based face forgery detection methods use convolutional neural networks (CNNs) [23,21,9,25,24,20] or recurrent neural networks (RNNs) [18,26] for detection. Former focuses solely on spatial information, while later con-siders temporal features as well.…”

Section: Face Forgery Detectionmentioning

confidence: 99%

Frequency domain deepfake detection based on two-stream neural network

Zhang

Sun

2023

Fourteenth International Conference on Graphics and Image Processing (ICGIP 2022)

View full text Add to dashboard Cite

Benefiting from the progress of deep learning driven generative models, face forgery technologies have rapidly become mature, thus raising public concerns about the illegal usage of these technologies. Despite the fact that fake images and videos are often unrecognizable to human eyes, recent work has found that hidden artifacts can be exposed in frequency domain. Our meth-od introduces the idea of separating human face area using landmarks before mining the forgery patterns in frequency domain. This we believe can help the network learn more discriminative features, also, a two-stream learning framework combining single-frame pathway and multi-frame pathway is developed to mine frequency clues. Compared with previous methods, our approach showed good results while using a few training data and little time. The effectiveness of our method is shown on different versions of the FF++ and Celeb-DF dataset.

show abstract

“…Specifically, if the true overlap in pixels for < a, a + > and < a, a − > is o + and o − , the flexible margin m f lex is shown in (2), where d is the patch dimension. Then the flexible margin ranking loss L f lex is calculated as (3).…”

Section: Lossmentioning

confidence: 99%

“…Prevalent manifestions of misinformation include fake news, digitally manipulated images, deepfake videos and more. Efforts towards the development of automated detection methods for fake news [1,2,3], natural-image forensics [4,5,6] and deepfakes [7,8,9] have gained traction. However, an important yet almost neglected field is that of biomedical image forensics.…”

Section: Introductionmentioning

confidence: 99%

MONet: Multi-scale Overlap Network for Duplication Detection in Biomedical Images

Sabir¹,

Nandi²,

AbdAlmageed³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Manipulation of biomedical images to misrepresent experimental results has plagued the biomedical community for a while. Recent interest in the problem led to the curation of a dataset and associated tasks to promote the development of biomedical forensic methods. Of these, the largest manipulation detection task focuses on the detection of duplicated regions between images. Traditional computer-vision based forensic models trained on natural images are not designed to overcome the challenges presented by biomedical images. We propose a multi-scale overlap detection model to detect duplicated image regions. Our model is structured to find duplication hierarchically, so as to reduce the number of patch operations. It achieves state-of-the-art performance overall and on multiple biomedical image categories.

show abstract

MEG: Multi-Evidence GNN for Multimodal Semantic Forensics

Cited by 10 publications

References 19 publications

Deepfake Video Detection Based on MesoNet with Preprocessing Module

Deepfake Video Detection Based on MesoNet with Preprocessing Module

Frequency domain deepfake detection based on two-stream neural network

MONet: Multi-scale Overlap Network for Duplication Detection in Biomedical Images

Contact Info

Product

Resources

About