SOTAVerified

DeepFake Detection

DeepFake Detection is the task of detecting fake videos or images that have been generated using deep learning techniques. Deepfakes are created by using machine learning algorithms to manipulate or replace parts of an original video or image, such as the face of a person. The goal of deepfake detection is to identify such manipulations and distinguish them from real videos or images.

Description source: DeepFakes: a New Threat to Face Recognition? Assessment and Detection

Image source: DeepFakes: a New Threat to Face Recognition? Assessment and Detection

Papers

Showing 501525 of 580 papers

TitleStatusHype
FFR_FD: Effective and Fast Detection of DeepFakes Based on Feature Point DefectsCode1
Understanding the Security of Deepfake Detection0
Automated Deepfake Detection0
Reverse Engineering of Generative Models: Inferring Model Hyperparameters from Generated ImagesCode1
DFGC 2021: A DeepFake Game CompetitionCode1
FReTAL: Generalizing Deepfake Detection using Knowledge Distillation and Representation Learning0
Deepfake Detection by Human Crowds, Machines, and Machine-informed CrowdsCode0
TAR: Generalized Forensic Framework to Detect Deepfakes using Weakly Supervised LearningCode0
What's wrong with this video? Comparing Explainers for Deepfake Detection0
One Shot Face Swapping on MegapixelsCode1
An Examination of Fairness of AI Models for Deepfake Detection0
DeepfakeUCL: Deepfake Detection via Unsupervised Contrastive Learning0
M2TR: Multi-modal Multi-scale Transformers for Deepfake DetectionCode1
Contrastive Learning of Global-Local Video RepresentationsCode1
Towards Measuring Fairness in AI: the Casual Conversations Dataset0
Deepfake Detection Scheme Based on Vision Transformer and DistillationCode1
Deepfake Forensics via An Adversarial GameCode1
KoDF: A Large-scale Korean DeepFake Detection Dataset0
DefakeHop: A Light-Weight High-Performance Deepfake DetectorCode1
Deepfake Videos in the Wild: Analysis and DetectionCode1
Multi-attentional Deepfake DetectionCode1
DeepFake-o-meter: An Open Platform for DeepFake Detection0
Countering Malicious DeepFakes: Survey, Battleground, and HorizonCode1
Deepfakes Generation and Detection: State-of-the-art, open challenges, countermeasures, and way forward0
Deepfake Video Detection Using Convolutional Vision TransformerCode1
Show:102550
← PrevPage 21 of 24Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AV-Lip-Sync+Accuracy (%)99.29Unverified
2AvtenetAccuracy (%)98.57Unverified
3FACTORROC AUC97.4Unverified
4RealForensicsROC AUC97.1Unverified
5AVADROC AUC94.5Unverified
6AV-Lip-Sync ModelAccuracy (%)94Unverified
7FTCNROC AUC93.1Unverified
8LipForensicsROC AUC91.1Unverified
9Multimodal Ensemble ModelAccuracy (%)89Unverified
10AD DFDROC AUC88.1Unverified
#ModelMetricClaimedVerifiedStatus
1XceptionNetDF96.36Unverified
2QAD-EAUC0.96Unverified
3EfficientNetB4 + EfficientNetB4ST + B4Att + B4AttSTAUC0.94Unverified
4MARLIN (ViT-L)AUC0.94Unverified
5MARLIN (ViT-B)AUC0.93Unverified
6MARLIN (ViT-S)AUC0.89Unverified
7EfficientNetB4 + EfficientNetB4ST + B4AttSTLogLoss0.33Unverified
#ModelMetricClaimedVerifiedStatus
1Cross Efficient Vision TransformerAUC0.95Unverified
2Efficient Vision TransformerAUC0.92Unverified
3EfficientNetB4 + EfficientNetB4ST + B4AttLogLoss0.46Unverified
#ModelMetricClaimedVerifiedStatus
1STYLE0L99Unverified
#ModelMetricClaimedVerifiedStatus
1FasterThanLiesAUC99.65Unverified
#ModelMetricClaimedVerifiedStatus
1FasterThanLiesAUC1Unverified
#ModelMetricClaimedVerifiedStatus
1FasterThanLiesAUC1Unverified
#ModelMetricClaimedVerifiedStatus
1BA-TFDAUC0.99Unverified