Metadata-Version: 2.4
Name: deepclip
Version: 0.1.0
Summary: CLIP for deepfakes: face forgery detection aligned with CLIP text embeddings
Author-email: seoyunje <seoyunje2001@gmail.com>
License-Expression: MIT
Project-URL: Homepage, https://github.com/SEOYUNJE/DeepClip
Project-URL: Bug Tracker, https://github.com/SEOYUNJE/DeepClip/issues
Project-URL: Source Code, https://github.com/SEOYUNJE/DeepClip
Project-URL: Releases, https://github.com/SEOYUNJE/DeepClip/releases
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch>=2.4.0
Requires-Dist: timm>=1.0.0
Requires-Dist: torchmetrics>=1.4.0
Requires-Dist: albumentations>=2.0.8
Requires-Dist: opencv-python>=5.0.0.93
Requires-Dist: numpy<3.0.0,>=1.23.0
Requires-Dist: pandas>=2.2.3
Requires-Dist: scikit-learn>=1.6.1
Requires-Dist: regex>=2023.0.0
Dynamic: license-file

# DeepClip
CLIP for deepfakes

## Dataset

[`preprocess/`](preprocess/README.md) builds the frame-level FaceForensics++ dataset and its forgery annotation
(face crops, 68 facial landmarks, forgery masks and forgery features) used to train DeepClip.
