DeepFake Video Dataset整理
無肖像權可用資料集待整理
Scientists warn deepfakes are about to become undetectable
https://www.popsci.com/technology/deepfakes-undetectable-heartbeats/
https://github.com/OpenTAI/wild-deepfake
https://arxiv.org/pdf/2101.01456v1
曾在2026/6/9被使用於以下論文
Advancing Deepfake Video Detection with Deep Learning Based Multi-Stream Architecture
https://ieeexplore.ieee.org/document/11546967
The DeepFake Detection Challenge (DFDC) Dataset
https://arxiv.org/abs/2006.07397
https://ai.meta.com/datasets/dfdc/
https://www.innovatiana.com/en/datasets/deep-fake-detection-dfd-entire-original-dataset
曾在2026/6/11被使用於以下論文
FedMaxAdapt: Gradient-Guided Adaptive Sparsity for Scalable Multi-Modal Federated Deepfake Detection
https://ieeexplore.ieee.org/document/11551901
曾在2026/6/8被使用於以下論文(但這篇也有用到FaceForensics++ 與 Celeb-DF)
Deepfake Video Forensics Using ResNeXt–LSTM Visual Modeling and DSP-Enhanced Audio Spectrogram Analysis
https://ieeexplore.ieee.org/document/11547579
曾在2026/6/8被使用於以下論文
A Reliable and Accurate Face Swap Deepfake video Detection using Multi - Model AI/ML system
https://ieeexplore.ieee.org/document/11544874
KoDf: A large-scale korean deepfake detection dataset
https://openaccess.thecvf.com/content/WACV2026/papers/Yermakov_Deepfake_Detection_that_Generalizes_Across_Benchmarks_WACV_2026_paper.pdf
https://www.computer.org/csdl/proceedings-article/wacv/2026/551100a773/2ggP4UMfX8s
曾在2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
A-V Representation Learning via Audio Shift Prediction for Multimodal Deepfake Detection and Temporal Localization
https://openaccess.thecvf.com/content/WACV2026/papers/Anshul_A-V_Representation_Learning_via_Audio_Shift_Prediction_for_Multimodal_Deepfake_WACV_2026_paper.pdf
https://dspace.iiti.ac.in/handle/123456789/16794
https://ieeexplore.ieee.org/document/11099399
2025年剛出不久，由164 位印度裔受試者，在收集任何臉部資料之前，皆已取得每位參與者的知情同意。目前2026年尚未有IEEE論文引用。
https://arxiv.org/abs/2108.05080
(論文原始有提到資料收集有涵蓋到youtube抓取)
MIS-AVoiDD: Modality Invariant and Specific Representation for Audio-Visual Deepfake Detection
https://ieeexplore.ieee.org/document/10460064 ->這篇也有用到KoDF
https://arxiv.org/abs/1909.12962
https://github.com/yuezunli/celeb-deepfakeforensics
曾在2026/6/8被使用於以下論文
A YOLO - MViT Based Hybrid Approach for Deepfake Videos Detection Using Facial Features
https://ieeexplore.ieee.org/document/11544639
DeeperForensics-1.0
https://arxiv.org/abs/2001.03024
邀請了 100 位受酬演員並蒐集高解析度（1920×1080）來源資料，涵蓋各種姿勢、表情與光照。包含多達 60,000支高品質影片，已取得所有演員的同意，以使用並操作其臉部。
曾在2026/4/30被使用於以下論文
EfficientNetB7 & Bi-LSTM for Deepfake Detection
https://ieeexplore.ieee.org/document/11486188
DeepfakeTIMIT
https://arxiv.org/abs/1812.08685
https://conradsanderson.id.au/vidtimit/
https://yinanhe.github.io/projects/forgerynet.html#Video
(論文原始有提到資料收集有涵蓋到youtube抓取)
https://arxiv.org/abs/2103.16076
(論文原始有提到資料收集有涵蓋到youtube抓取)
DARK FACE: Face Detection in Low Light Condition
https://flyywh.github.io/CVPRW2019LowLight/
Video Deepfake Detection using Remote Photoplethysmography Feature and Machine Learning
https://www.csie.ntpu.edu.tw/uploads/file/project/111/PRJ-NTPUCSIE-111-004.pdf
Daniel, M., & Ethan, B. (2019). iPhys: An Open Non-Contact Imaging-Based Physiological Measurement Toolbox. arXiv preprint arXiv:1901.04366 [cs.CV].
https://arxiv.org/abs/1901.04366
https://github.com/danmcduff/iphys-toolbox
TWHD:台灣開拓性的多臉偽造檢測深偽資料集
TWHD:Taiwan Pioneering Deepfake Dataset for Multi-Face Forgery Detection
https://hdl.handle.net/11296/8u83s3
留言
張貼留言