README.md

---
license: mit
pipeline_tag: video-classification
tags:
  - deepfake-detection
  - face-forensics
  - person-of-interest
  - masked-autoencoder
  - arxiv:2606.20302
---

# CUPID — Person-of-Interest Deepfake Detection

Weights for [CUPID](https://github.com/polimi-ispl/CUPID): *Reconstructing UV
Texture Maps for Interpretable Person-of-Interest Deepfake Detection*.

📄 **Paper:** [arXiv:2606.20302](https://arxiv.org/abs/2606.20302)

`cupid_mae.pth` is a ViT-Tiny Masked Autoencoder trained self-supervised on UV
face textures of real videos (VoxCeleb2). The CUPID pipeline scores a test
video by max cosine similarity between its CLS-token features and those of
reference videos of the person of interest.

## Usage

```bash
pip install git+https://github.com/polimi-ispl/CUPID
cupid extract-reference --reference ref1.mp4 ref2.mp4 -o poi.pt
cupid score --reference-set poi.pt --test test.mp4
```

Weights are downloaded automatically on first use.

## Third-party weights

CUPID's UV-texture extraction uses four asset files from
[3DDFA_V3](https://github.com/wang-zidu/3DDFA-V3) (CVPR 2024), which are NOT
mirrored here. They are downloaded directly from the authors' repository at
[Zidu-Wang/3DDFA-V3](https://huggingface.co/datasets/Zidu-Wang/3DDFA-V3) and
remain subject to their respective licenses and provenance (RetinaFace
weights from biubug6/Pytorch_Retinaface, large_base_net.pth from HRN,
net_recon.pth from 3DDFA_V3, face_model.npy derived from the Basel Face
Model, Exp_Pca, and Deep3D).

## License

The CUPID checkpoint is released under the MIT license.

## Citation

```bibtex
@article{affatato2026cupid,
  title   = {{CUPID}: Reconstructing UV Texture Maps for Interpretable
             Person-of-Interest Deepfake Detection},
  author  = {Affatato, Giovanni and Mandelli, Sara and Bestagini, Paolo and Tubaro, Stefano},
  journal = {arXiv preprint arXiv:2606.20302},
  year    = {2026},
}
```