
prithivMLmods/Deepfake-QualityAssess-88M
Image Classification
•
Updated
•
55
•
7
image
imagewidth (px) 512
3.31k
| label
class label 2
classes |
---|---|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
|
0Bad
|
Deepfake QA is a Deepfake Quality Assessment model designed to analyze the quality of deepfake images & videos. It evaluates whether a deepfake is of good or bad quality, where:
This classification serves as the foundation for training models on deepfake quality assessment, helping improve deepfake detection and enhancement techniques.
If you use our dataset or models, please cite:
@misc{deepfake_quality_assessment_2025,
author = {Wildy AI Team Collaborations},
title = {Deepfake Quality Assessment Models},
year = {2025},
note = {Early release},
models_training = {@prithivMLmods},
dataset_curation_strategy = {@prithivMLmods},
dataset_curation = {Wildy AI Team}
}