license: mit
tags:
- not-for-all-audiences
- text-to-image
- stable-diffusion
size_categories:
- n<1K
DFLIP-3K raw data:
This is the raw data of DFLIP-3K, a deepfake database (DFLIP-3K) for the development of convincing and explainable deepfake detection.
How to use:
The training set contains data that we may use during training (as we only use a portion of the data), while the test set is used for testing purposes. The data may include factors such as pornography, gore, and other sensitive content. Due to technical constraints, we are unable to conduct a comprehensive manual review of the entire dataset. Please use the data with caution. By using this dataset, you agree to the below terms and conditions.
Data Set Usage Agreement
This Data Set Usage Agreement (hereinafter referred to as "the Agreement") is a legal document between you (hereinafter referred to as "Licensee") and the data set provider, concerning the use of the data set provided. Please read the following terms and conditions carefully before using the data set.
Definitions
- "Data Set" refers to the data set obtained by the data set provider through web scraping techniques and the partial data extracted from related works.
- "Data Set Provider" refers to the individual, company, or organization providing this data set.
- "Licensee" refers to any individual or entity that accepts the terms of this agreement to use the data set for academic research.
- "Offensive Information" refers to content that may be considered disrespectful, discriminatory, aggressive, or otherwise inappropriate.
Scope of License
- The data set provider grants the Licensee a non-exclusive, non-transferable, and non-distributable license to use the data set solely for academic research purposes.
- The Licensee shall not distribute, sell, transfer, share, or otherwise make available the data set or any part thereof to any third party.
- The Licensee must refer to the works cited in the data set to obtain the integrity and background information of the data set.
Disclaimer
- The data set may contain Offensive Information, and the data set provider is not liable for such content.
- The data set is provided "as is," and the data set provider makes no express or implied warranties of accuracy, completeness, or fitness for a particular purpose.
Limitation of Liability
- The data set provider shall not be liable for any direct, indirect, incidental, special, consequential, or punitive damages that may result from the use of the data set by the Licensee or any third party.
- The Licensee agrees to indemnify and hold the data set provider harmless from any claims, damages, liabilities, costs, and expenses (including reasonable attorneys' fees) that may arise from any third party due to the use of the data set.
Intellectual Property
- The data set and all related intellectual property rights are owned by the data set provider or their licensors.
- This agreement does not grant the Licensee any rights to intellectual property, except as expressly provided.
Termination
- If the Licensee breaches any term or condition of this Agreement, it shall automatically terminate, and the Licensee must immediately cease using the data set and destroy all copies of the data set.
General Terms
- This Agreement constitutes the entire agreement between the parties regarding the use of the data set and supersedes all prior and contemporaneous agreements, commitments, and understandings.
- If any part of this Agreement is found to be invalid or unenforceable, the remaining parts will remain in effect.
- The interpretation and enforcement of this Agreement shall be governed by the laws of the jurisdiction where the data set provider is located.
By using the data set, the Licensee acknowledges that they have read and understood the terms of this Agreement and agree to be bound by them.
Acknowledgements:
Part of the data used in this work comes from the following projects. Please refer to these projects for more detailed information:
https://huggingface.co/datasets/poloclub/diffusiondb
https://huggingface.co/datasets/wanng/midjourney-v5-202304-clean
Cite:
@article{wang2024linguistic, title={Linguistic Profiling of Deepfakes: An Open Database for Next-Generation Deepfake Detection}, author={Wang, Yabin and Huang, Zhiwu and Ma, Zhiheng and Hong, Xiaopeng}, journal={arXiv preprint arXiv:2401.02335}, year={2024} }