Diffusion-Inspired Dynamic Models for Enhanced Fake Face Detection

Mamadou Bah, Rajjeshwar Ganguly, Mohamed Dahmane

2024

Abstract

The conventional convolutional U-Net model was originally designed to segment images. Enhanced versions of the architecture with timestep embedding and self-attention inspired by transformers were proposed in the literature for image classification tasks. In this paper, we investigate a U-Net–encoder architecture for deepfake detection that involves two key features, the use of self-attention blocks to capture both local and global content representation, and the integration of timestep embedding to capture the dynamic perturbation of the input data. The model is trained and evaluated on FF++ dataset, comprising of real and deepfake synthesized videos. Notably, compared to traditional models pretrained on ImageNet, our model demonstrates superior performance. The experimental results highlight the effectiveness of our approach in achieving improved classification results for the challenging task of distinguishing real and deepfake images. The achieved performances suggest that the model aims to leverage both spatial information and dynamic perturbation for improved detection performance.

Paper Citation

in Harvard Style

Bah M., Ganguly R. and Dahmane M. (2024). Diffusion-Inspired Dynamic Models for Enhanced Fake Face Detection. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 532-538. DOI: 10.5220/0012391300003660

in Bibtex Style

@conference{visapp24,
author={Mamadou Bah and Rajjeshwar Ganguly and Mohamed Dahmane},
title={Diffusion-Inspired Dynamic Models for Enhanced Fake Face Detection},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP},
year={2024},
pages={532-538},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012391300003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP
TI - Diffusion-Inspired Dynamic Models for Enhanced Fake Face Detection
SN - 978-989-758-679-8
AU - Bah M.
AU - Ganguly R.
AU - Dahmane M.
PY - 2024
SP - 532
EP - 538
DO - 10.5220/0012391300003660
PB - SciTePress