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

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 51–100 of 580 papers

TitleStatusHype
AWARE-NET: Adaptive Weighted Averaging for Robust Ensemble Network in Deepfake DetectionCode0
End-to-end Audio Deepfake Detection from RAW Waveforms: a RawNet-Based Approach with Cross-Dataset EvaluationCode1
Towards Generalizable Deepfake Detection with Spatial-Frequency Collaborative Learning and Hierarchical Cross-Modal Fusion—0
Benchmarking Audio Deepfake Detection Robustness in Real-world Communication Scenarios—0
Big Brother is Watching: Proactive Deepfake Detection via Learnable Hidden Face—0
FractalForensics: Proactive Deepfake Detection and Localization via Fractal Watermarks—0
LoRAX: LoRA eXpandable Networks for Continual Synthetic Image AttributionCode0
Detect All-Type Deepfake Audio: Wavelet Prompt Tuning for Enhanced Auditory PerceptionCode1
Face-LLaVA: Facial Expression and Attribute Understanding through Instruction Tuning—0
Nes2Net: A Lightweight Nested Architecture for Foundation Model Driven Speech Anti-spoofingCode2
Comparative Analysis of Deepfake Detection Models: New Approaches and Perspectives—0
Synthetic Art Generation and DeepFake Detection A Study on Jamini Roy Inspired Dataset—0
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face DetectorCode2
Can Multi-modal (reasoning) LLMs work as deepfake detectors?—0
Unlocking the Hidden Potential of CLIP in Generalizable Deepfake DetectionCode2
Anomaly Detection and Localization for Speech Deepfakes via Feature Pyramid Matching—0
Measuring the Robustness of Audio Deepfake DetectorsCode0
TruthLens: Explainable DeepFake Detection for Face Manipulated and Fully Synthetic Data—0
Unlocking the Capabilities of Vision-Language Models for Generalizable and Explainable Deepfake Detection—0
TruthLens:A Training-Free Paradigm for DeepFake Detection—0
ExDDV: A New Dataset for Explainable Deepfake Detection in VideoCode1
Deepfake Detection with Optimized Hybrid Model: EAR Biometric Descriptor via Improved RCNN—0
Deepfake Detection of Face Images based on a Convolutional Neural Network—0
VoD: Learning Volume of Differences for Video-Based Deepfake DetectionCode0
VLForgery Face Triad: Detection, Localization and Attribution via Multimodal Large Language Models—0
Reduced Spatial Dependency for More General Video-level Deepfake Detection—0
Deepfake Detection via Knowledge Injection—0
Deepfake-Eval-2024: A Multi-Modal In-the-Wild Benchmark of Deepfakes Circulated in 2024Code1
DeePen: Penetration Testing for Audio Deepfake Detection—0
Methods and Trends in Detecting Generated Images: A Comprehensive Review—0
Lightweight Deepfake Detection Based on Multi-Feature Fusion—0
Do Deepfake Detectors Work in Reality?—0
CAE-Net: Generalized Deepfake Image Detection using Convolution and Attention Mechanisms with Spatial and Frequency Domain Features—0
Generalizable speech deepfake detection via meta-learned LoRA—0
A Preliminary Exploration with GPT-4o Voice Mode—0
Exploring Active Data Selection Strategies for Continuous Training in Deepfake Detection—0
Improving the Perturbation-Based Explanation of Deepfake Detectors Through the Use of Adversarially-Generated SamplesCode0
REALEDIT: Reddit Edits As a Large-scale Empirical Dataset for Image Transformations—0
Comprehensive Layer-wise Analysis of SSL Models for Audio Deepfake DetectionCode1
Deepfake Detection of Singing Voices With Whisper Encodings—0
DFCon: Attention-Driven Supervised Contrastive Learning for Robust Deepfake Detection—0
Extending Information Bottleneck Attribution to Video SequencesCode0
Classifying Deepfakes Using Swin Transformers—0
Generalizable Deepfake Detection via Effective Local-Global Feature ExtractionCode1
Generalizable Audio Deepfake Detection via Latent Space Refinement and Augmentation—0
GC-ConsFlow: Leveraging Optical Flow Residuals and Global Context for Robust Deepfake Detection—0
What Does an Audio Deepfake Detector Focus on? A Study in the Time Domain—0
Audio-Visual Deepfake Detection With Local Temporal Inconsistencies—0
Neural Codec Source Tracing: Toward Comprehensive Attribution in Open-Set ConditionCode0
HFMF: Hierarchical Fusion Meets Multi-Stream Models for Deepfake DetectionCode1
Show:102550
← PrevPage 2 of 12Next →
All datasetsFakeAVCelebFaceForensicsDFDC1CIFAKE: Real and AI-Generated Synthetic ImagesCOCOFakeDFFDLAV-DF

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AV-Lip-Sync+Accuracy (%)99.29—Unverified
2AvtenetAccuracy (%)98.57—Unverified
3FACTORROC AUC97.4—Unverified
4RealForensicsROC AUC97.1—Unverified
5AVADROC AUC94.5—Unverified
6AV-Lip-Sync ModelAccuracy (%)94—Unverified
7FTCNROC AUC93.1—Unverified
8LipForensicsROC AUC91.1—Unverified
9Multimodal Ensemble ModelAccuracy (%)89—Unverified
10AD DFDROC AUC88.1—Unverified
#ModelMetricClaimedVerifiedStatus
1XceptionNetDF96.36—Unverified
2QAD-EAUC0.96—Unverified
3EfficientNetB4 + EfficientNetB4ST + B4Att + B4AttSTAUC0.94—Unverified
4MARLIN (ViT-L)AUC0.94—Unverified
5MARLIN (ViT-B)AUC0.93—Unverified
6MARLIN (ViT-S)AUC0.89—Unverified
7EfficientNetB4 + EfficientNetB4ST + B4AttSTLogLoss0.33—Unverified
#ModelMetricClaimedVerifiedStatus
1Cross Efficient Vision TransformerAUC0.95—Unverified
2Efficient Vision TransformerAUC0.92—Unverified
3EfficientNetB4 + EfficientNetB4ST + B4AttLogLoss0.46—Unverified
#ModelMetricClaimedVerifiedStatus
1STYLE0L99—Unverified
#ModelMetricClaimedVerifiedStatus
1FasterThanLiesAUC99.65—Unverified
#ModelMetricClaimedVerifiedStatus
1FasterThanLiesAUC1—Unverified
#ModelMetricClaimedVerifiedStatus
1FasterThanLiesAUC1—Unverified
#ModelMetricClaimedVerifiedStatus
1BA-TFDAUC0.99—Unverified