If you want to access the DeepMoiréFake dataset for download, please follow this license agreement.
Note that If you use our dataset, please consider using the citation.
@inproceedings{
tariq2025through,
title={Through the Lens: Benchmarking Deepfake Detectors Against Moir\'e-Induced Distortions},
author={Razaib Tariq and Minji Heo and Shahroz Tariq and Simon S. Woo},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
year={2025},
url={https://openreview.net/forum?id=HtGOaraQEH}}
"description": "DeepMoireFake is an real-world environmental dataset that influences deepfake detectors.",
"keywords": [
"Computer and Information Science",
"Deepfake Detection",
"Distortion Artifacts",
"Deepfake",
"Distortion",
"Moire",
"Synthetic",
"Compression"
],
"license": "https://dataverse.harvard.edu/api/datasets/:persistentId/versions/2.0/customlicense?persistentId=doi:10.7910/DVN/XYOSYW",
"datePublished": "2025-05-16",
"dateModified": "2025-05-21",
"includedInDataCatalog": {
"@type": "DataCatalog",
"name": "Harvard Dataverse",
"url": "https://dataverse.harvard.edu"
},
"publisher": {
"@type": "Organization",
"name": "Harvard Dataverse"
},
"version": "2.0",
"citeAs": "@data{DVN/XYOSYW_2025,author = {Tariq, Razaib and Heo, Minji and Simon S.Woo and Tariq, Shahroz},publisher = {Harvard Dataverse},title = {Through the Lens: Benchmarking Deepfake Detectors Against Moiré-Induced Distortions},year = {2025},url = {https://doi.org/10.7910/DVN/XYOSYW}}",
"distribution": [
{
"@type": "cr:FileObject",
"@id": "deepmoirefake.part001.rar",
"name": "deepmoirefake.part001.rar",
"encodingFormat": "application/x-rar-compressed",
"md5": "f22ab4d37b2b221b1953e67a64653e74",
"contentSize": "2147483648",
"description": "",
"contentUrl": "https://dataverse.harvard.edu/api/access/datafile/11237743"
}, .......
{
"@type": "cr:FileObject",
"@id": "deepmoirefake.part101.rar",
"name": "deepmoirefake.part101.rar",
"encodingFormat": "application/x-rar-compressed",
"md5": "96e34de322d41a21ad8b244c2293a825",
"contentSize": "687418651",
"description": "",
"contentUrl": "https://dataverse.harvard.edu/api/access/datafile/11551159"
}
📱 Samsung
Lights on/off
🖥️ Screen
LG
CelebDF
Real
Fake
DFDC
Real
Fake
DFD
Real
Fake
FaceForensics++
Deepfake
Real
Fake
Face2Face
Real
Fake
Faceswap
Real
Fake
Neuraltextures
Real
Fake
UADFV
Real
Fake
Benq
.....
Samsung
.....
Lenovo
.....
📱 iPhone
Lights 🌇 on/off 🌆
🖥️ Screen
LG
CelebDF
Real
Fake
DFDC
Real
Fake
DFD
Real
Fake
FaceForensics++
Deepfake
Real
Fake
Face2Face
Real
Fake
Faceswap
Real
Fake
Neuraltextures
Real
Fake
UADFV
Real
Fake
Benq
.....
Samsung
.....
Lenovo
.....
🧾 Release History
Please send an access request to download the dataset: Dataverse Harvard
Our DeepMoiréFake dataset has a digital object identifier (DOI) as follows: DeepMoireFake DOI
Our Eval codes are here: https://github.com/Razaib-Tariq/DeepMoireFake
Dataset samples can be found here: https://github.com/Razaib-Tariq/DeepMoireFake/tree/main/Samples