This repository provides a list a list of publications from DASHLab about Deepfakes.
Our published datasets of deepfakes.
2023: Towards Understanding of Deepfake Videos in the Wild accepted by CIKM23. [paper]
2022: Zoom-DF: A dataset for Video Conferencing Deepfake accepted by WDC22. [paper ]
2021: FakeAVCeleb: A Novel Audio-Video Multimodal Deepfake Dataset accepted by NeurIPS21. [paper]
Methods to generalize deepfake detectors.
2021: One Detector to Rule Them All: Towards a General Deepfake Attack Detection Framework accepted by WWW21. [paper]
2021: FReTAL: Generalizing Deepfake Detection using Knowledge Distillation and Representation Learning accepted by CVPRW21. [paper]
2021: TAR: Generalized Forensic Framework to Detect Deepfakes using Weakly Supervised Learning accepted by IFIP-SEC 21. [paper]
Our published papers that related to detecting low-quality compressed deepfakes.
2023: Quality-Agnostic Deepfake Detection with Intra-model Collaborative Learning accepted by ICCV23. [paper]
2022: ADD: Frequency Attention and Multi-View Based Knowledge Distillation to Detect Low-Quality Compressed Deepfake Images accepted by AAAI22. [paper]
2022: BZNet: Unsupervised Multi-scale Branch Zooming Network for Detecting Low-quality Deepfake Videos accepted by WWW22. [paper]
Our published papers that related to applying continual learning to detect newly arrived deepfakes/GAN.
2021: CoReD: Generalizing Fake Media Detection with Continual Representation using Distillation accepted by ACMMM21. [paper]
2020: T-GD: Transferable GAN-generated Images Detection Framework accepted by ICML20. [paper]
Our published papers that related to evasion techniques to fool deepfake detectors.
2024: Beyond the Screen: Evaluating Deepfake Detectors under Moir ́e Pattern Effects accepted by WMF-CVPR24. [paper]
2024: Exploring the Impact of Moiré Pattern on Deepfake Detectors accepted by ICIP24. [paper]
2023: Evaluating Racial Bias in Face Recognition APIs using Deepfakes accepted by IEEE Computer Magazine.
2022: Am I a Real or Fake Celebrity? Evaluating Face Recognition and Verification APIs under Deepfake Impersonation Attack accepted by WWW2022. [paper]
2022: Evading Deepfake Detectors via High-Quality Face Pre-Processing Methods accepted by ICPR22. [paper]
2022: Negative Adversarial Example Generation Against Naver’s Celebrity Recognition API accepted by WDC22. [paper]
2022: A Face Pre-Processing Approach to Evade Deepfade Detector accepted by WDC22. [paper]
Our published papers that related to applying multi-model to detect deepfakes.
2022: Evaluation of an Audio-Video Multimodal Deepfake Dataset using Unimodal and Multimodal Detectors accepted by ADGD21. [paper]
2024: UGAD: Universal Generative AI Detector utilizing Frequency Fingerprints accepted by CIKM24*. [paper]
2024: iFakeDetector: Real Time Integrated Web-based Deepfake Detection System accepted by IJCAI24*. [paper]
2023: SAFE: Sequential Attentive Face Embedding with Contrastive Learning for Deepfake Video Detection accepted by CIKM23*. [paper]
2022: Deepfake Detection for Facial Images with Facemasks accepted by WDC22. [paper]
2021: Exploring the Asynchronous of the Frequency Spectra of GAN-generated Facial Images accepted by IJCAIW21. [paper]
2020: OC-FakeDect: Classifying Deepfakes Using One-class Variational Autoencoder accepted by CVPRW20. [paper]
2020: FDFtNet: Facing Off Fake Images using Fake DetectionFine-tuning Network accepted by IFIP-SEC 20. [paper]
2019: GAN is a Friend or Foe? A Framework to Detect Various Fake Face Images accepted by ACMSAC19. [paper]
2023-2025. 악의적 변조 콘텐츠 대응을 위한 딥페이크 탐지 고도화, 생성억제, 유포 방지 플랫폼 개발 (IITP)
2022-2023. 딥페이크 탐지 및 칩페이크 탐지 모델 연구 및 고도화 (삼성SDS)
2020-2022. 인공지능 기반의 딥페이크 멀티미디어 탐지 연구 (한국연구재단 우수신진)
2020. 지능형 동영상 합성(딥페이크) 검출을 위한 연구 용역 (국립과학수사연구원)
2020. 딥페이크 기술을 활용한 온라인 불법행위 대응방안 (경찰청)